Sugar Research and Development Corporation Final Report Project No CSR22 Project Title: Best-practice irrigation management to maximise profitability and ensure sustainability in the Ord Sugar Industry SRDC Program: Program 2: Crop Management Organisations: CSRSugar Agriculture Western Australia CSIRO Sustainable Ecosystems Chief Investigator: DrA WWood Location: CSR Sugar, Technical Field Department, Macknade MiII, PMB4, Ingham Qld 4850 Telephone: 074777 0226 Facsimile: 074777 7656 Email: awood@csr.com.au Commencement Date: 1 July 1996 Completion Date: 1 October 2003 2 CONTENTS Best practice irrigation management Page No Executive Summary 5 1. Introduction and project background 7 2. Project objectives 11 3. Parties involved and links with other initiatives 12 4. Methodology 13 5. Benchmarking growers' irrigation practices 14 6. Drying off experiments 21 7. Irrigation scheduling trials 29 8. Water extraction by sugarcane on different soils 49 9. Weather conditions in the Ord for the duration ofthe project 63 10. Developing irrigation guidelines using the APSIM Sugarcane model 68 11. Irrigation water application efficiency 81 12. Irrigation guidelines 102 13. Second survey of irrigation practices 105 14. General discussion and conclusions 106 15. References 110 Analysis of outputs and outcomes compared with objectives 114 Potential benefits and likely impact for the Australian Sugar Industry 116 Recommended future research needs 117 Extension activities 119 Acknowledgements 124 3 Industry Development Officer Introduction, activities, outputs and industry outcomes Appendices Appendix 1. Pnblications, Reports and Extension Notes Appendix 2. Irrigation Scheduling 'Ute Guide' Appendix 3. Industry Development Officer End of Season Reports Appendix 4. Industry Development Officer Field Day Notes 125 127 129 130 143 4 EXECUTIVE SUMMARY The Ord Sugar industry was established in 1995 with the commissioning of a sugar mill and the first full season of commercial production was in 1996. It is now a major industry in Stage I of the Ord Irrigation Area, currently occupying over 4000 hectares and producing around 60,000 tonnes of raw sugar annually for export. The climatic conditions in the Ord impose a high water requirement for sugarcane crops. Growers face a considerable challenge in meeting that requirement whilst maximising their profitability and minimising drainage losses and potential impacts on the environment. Groundwater levels have risen substantially in the 30 years since the start of irrigation in the Ord. Consequently the development of appropriate irrigation practices for sugarcane is a key requirement in the management of rising water tables and in the sustainability of Ord sugarcane production. The first step in developing best practice irrigation management was to gain a better understanding of the range of irrigation practices being used by the industry at the beginning of the project. A survey conducted to benchmark irrigation practices used on the 1995/96 sugarcane crop indicated very high rates of annual water application, making the development of irrigation practices that maximise profitability and minimise groundwater accessions a key priority for tlle Ord sugar industry. Withholding irrigation prior to harvest, termed drying-off, is often used to dry out fields and increase sugar content, although responses can be highly variable. To examine this concept in the Ord, treatments were imposed on 3 crops of sugarcane in 1997 and four in 1998 by withholding irrigation for various lengths of time prior to harvest. Results indicated minimal responses to drying-off for periods of up to 80 days, with no increase in sucrose content of the cane. A trial was established in 1998 to quantiJY the yield responses to different irrigation schedules and assist in the development of irrigation practices that not only maximise sugar yields but also improve the efficiency of irrigation water use and stabilise rising water tables. Surprisingly, there was little response to the treatments imposed which represented the extremes in likely irrigation schedules. The pattern of biomass accumulation in both the drying-off and irrigation scheduling trials indicated a slow-down in crop growth in the later stages of crop development. Comparisons with other areas showed tllat early grOwtll rates were not much lower than those measured in the Burdekin but that later growth was significantly slower. Measurements of crop water use confirmed that high rates of water use occurred before the wet season but much lower rates relative to evaporation occurred after the wet season. Clearly there appears to be a close relationship between the pattern of crop biomass accumulation and the rate of crop water use with a marked slow-down in both occurring following the onset of the wet season. This has important implications for irrigation water management guidelines with more frequent irrigations being appropriate during the hottest part of the year leading up to the wet season and much less frequent irrigations being needed after the wet season, which also appears to be a critical period for accessions to the water table. Measurements of water extraction on a range of soils indicated very high levels of plant extractable soil water. This provides additional support for much longer intervals between irrigations but possibly more water applied at each irrigation. Crop modelling using the APSIM Sugarcane model and the experimental data gathered in this project was used to develop irrigation guidelines for different conditions. The model had 5 to be reconfigured based on the slower growth rates during the later stages of crop development. Another model, the Surface Irrigation model (SIRMOD) was used to simulate the requirements for achieving optimum water application efficiency for furrow irrigation of the cane crop. Data collected from different furrow-irrigated bays of sugarcane were used to configure SIRMOD so that it simulated irrigation events over a range of different conditions in the Ord. It was used to develop 'look up' tables indicating expected irrigation application efficiencies when factors such as moisture deficit, furrow inflow rate, irrigation duration, furrow length and furrow shape were varied. These guidelines for optimising irrigation application efficiency are amongst the most comprehensive anywhere in the Australian sugar industry. Best practice irrigation guidelines, which maXImIse profitability and mimmise losses of water, are now available to the Ord sugar industry. They include recommendations on irrigation frequency, soil moisture at irrigation, tail water flow, furrow shape, optimum row length and management of deep drainage. However, there remains considerable scope for growers to make further improvements to their irrigation management practices and tailor their applications more closely to the specific requirements of their crops and soils, and it will not be until this occurs that regional water tables in the Ord will start to stabilise. The following benefits are likely to arise from this project and impact on the sugar industries in the Ord and in eastern Australia: • better matching of irrigation applications to crop requirements leading to reduced use of irrigation water and reduced accessions to groundwater; • improved likelihood of meeting future water licensing requirements and targets; • reduced losses of irrigation water through deep drainage and tailwater runoff; • longer viability of laser levelled cane blocks and reduced siltation of tailwater drains; • improved profitability with less frequent irrigations through reduced irrigation costs; • improved social impact through less time spent irrigating; • reduced pressure on irrigation infrastructure; • stabilisation of groundwater tables; • reducing pressures from environmental lobby groups; • application of knowledge gained from this project to other irrigated sugarcane areas such as the Burdekin and to other irrigated crops grown in the Ord. 6 1. INTRODUCTION AND PROJECT BACKGROUND The Ord River Irrigation Area The Ord River Irrigation Area (ORIA), which was developed in 1962 with the construction of the Diversion Dam which led to the formation of Lake Kununurra, comprises approximately 13,000 ha of irrigable soils. In 1973, the bnilding of the Ord Dam that created Lake Argyle ensured a water snpply for an irrigation project of more than 50,000 ha. The irrigation system was constrncted to take advantage of abnndant supplies of water with a network of channels fed by a single main channel delivering water to farms. Stage 1 of the ORiA consists of around 10,000 ha of land on the Ivanhoe Plain with mainly cracking clay soils of the Cununurra Series. Areas between the plain and the river are sandier soils of the river levee and the Pago and Cockatoo Sands between the plain and the nearby hill systems. Some areas of very heavy Aquitaine clay soils are located on the flatter margins of the main Ivanhoe Plain. A further 2000 ha of land on the Packsaddle Plain includes most of the soil types found on the Ivanhoe Plain as well as significant areas of medium textured red clay soils of the Packsaddle Series. More than 60% of the ORiA is underlain by sand and gravel beds from the old channel of the Ord River. These conduct water well and form extensive interconnected aquifers beneath the irrigated areas of both the Ivanhoe and Packsaddle plains. The water in these aquifers usually has low levels of dissolved salts. Following extensive cropping trials throughout the 1940's and 50's, the initial irrigation farming system was based on cropping cotton over the wet season. Farms of about 240 ha were released in the early 1960's and these were serviced with irrigation supply points and drainage outlets as required. Conditions were placed on the development of the farms to encourage rapid development of the irrigation land. In the 1960's and early 1970's, cotton was the main crop grown in the irrigation area. The cotton industry closed in 1974 and the area under crop declined significantly for a number of years. Towards the end of the 1980's horticulture was introduced and high value crops that could supply the domestic market out of season were adopted such as cucurbit crops, bananas and a range of small crops. Profits from these high value crops enabled other crops such as sugarcane to be grown and resulted in all of the available irrigation land being brought into production. The Ord Sugar Industry In late 1994 CSR Ltd., with the Ord River District Cooperative as a minor shareholder, commenced construction of a sugar mill. The sugar mill is a low cost, small capacity factory based on modern technology and low labour requirements. It was designed to be able to process a total crop of 560,000 tonnes at a crushing rate of 120 tOlmes cane per hour. The first cane was processed in late 1995 and the first full season of crushing was in 1996. This was the first commercial sugar industry established in the Ord even though sugarcane had been suggested as a suitable crop for many years. The system of sugarcane production employs furrow irrigation methods that are suitable given the low cost of irrigation water compared with other irrigated cane areas in Australia. The cane growers in the Ord were already established farmers growing a range of other irrigated crops and so furrow irrigated sugarcane fitted easily into the farming system. A strong history of highly efficient irrigation water use had not been established in the area as the provision of water had never been regarded as being limited or a significant cost. The sugar industry has not been able to achieve either the yield or CCS targets initially set in the planning phase and the production of high yielding and high CCS crops remains a 7 challenge for the industry to ensure its future viability. The expectation, when commercial sugarcane production began, was of large crops yielding in excess of 160 tonnes cane per hectare and with a high sugar content, comparable with the Burdekin region. Whilst cane yields have been highly variable but with some blocks reaching target yields the low sugar content and purity of Ord cane, particularly early in the season, has been the most surprising and disappointing aspect of commercial sugarcane production. Consequently a major reason for initiating the current research project was to examine different irrigation management strategies and find ways of achieving high cane and sugar yields in conjunction with efficient water use and negligible environmental impacts. Review ofprevious irrigatioll work conducted with sugarcane in the Ord Prior to the Ord development being funded, sugarcane had been identified as a potential crop for the region but repeated examination of sugar as a commercial crop and significant government and industry research commitments during the 1950's and 60's failed to result in a commercial industry (Gardiner, 1998). Early sugarcane research in the Ord River District occurred in three distinct periods: 1. The first period of work from 1950 - 1957 was conducted by the CSIRO at the Kimberley Research Station, Kununurra 2. In the second period from 1964 - 1968, CSR conducted further research into sugarcane production at Kimberley Reseach Station. 3. In the third period, from 1974 - 1982, work was conducted jointly by the Department of Agriculture and the Bureau of Sugar Experiment Stations on a commercial scale pilot farm. The first irrigation trials with sugarcane in the Ord were conducted by CSIRO in the 1950's at the Kimberley Research Station and were reported by Lee et al (1963). Based on measurements of cane growth in relation to moisture supply and visual observations of leaf water stress, Lee et al concluded that in the absence of rain an irrigation frequency of 10- I4 days was probably adequate for good cane growth. Subsequently CSR conducted an irrigation scheduling trial over a period of three years using the sugarcane varieties Trojan and Pindar. The four irrigation treatments in the plant crop were: 1. Irrigating every 10 days 2. Irrigating when the stalk extension of Pindar fell below 5mm/day ,3. Irrigating every three weeks 4. Irrigating when the crop appeared stressed. From the experimental data available in Gallagher (I 986), it is impossible to establish whether there were significant responses to the different irrigation treatments. Plant cane results showed no apparent differences in cane yield for the first three treatments but reduced yields for the fourth treatment. In first and second ratoons treatments were changed to: 1. Irrigating every 7-10 days 2. Irrigating every 10-14 days 3. Irrigating every 2-3 weeks 4. Irrigating when the crop appeared stressed. Again yields appeared to be lower in the fourth treatment. No apparent CCS differences occurred between any of the scheduling treatments imposed. Further research on sugarcane was conducted in the Ord from 1975-82 (Cox and Chapman, 1985), but with ample low priced irrigation water available little attention was paid to irrigation frequency or water management. Gallagher (1986) reported the results of an 8 irrigation trial with six different treatments established on the Pilot Farm using the sugarcane variety Trojan. In the plant crop the treatments included two different irrigation schedules based on different pan factors and further schedules were imposed in the first and second ratoon crops using a range of different pan factors. The results suggest that there were no apparent differences in cane yield or CCS. However the BSES report (BSES, 1982) mentions that the variety in this trial was badly affected by top rot. The conclusion from this trial work is that "the experiment contributed little towards providing preliminary recommendations for frequency of irrigating sugar cane on Cunnnurra clay soils" (Gallagher, 1986). Much of the early irrigation scheduling work with sugar in the Ord was inconclusive with little or no response to either irrigation frequency or to drying off strategies. Consequently the generally accepted schedules of 25-30 irrigations per crop year (watering every 10-14 days) were stilI the broad recommendation when the present commercial sugar industry began in 1994. Groundwater In common with most other irrigation areas throughout Australia and the rest of the world, groundwater levels have risen substantially in the 30 years since the start of irrigation in the ORIA and in some areas are so close to the surface that they require remedial measures. Groundwater at or near the surface in irrigation areas can reduce crop yields (and sugarcane in particular) not only through waterlogging but also through salinity in the root zone and sodicity from high sodium levels in groundwater causing soil structure problems. A good understanding of the effects of irrigation on the groundwater regime is therefore important for sustainable agriculture in the Ord River Irrigation Area (O'Boy et aZ, 2001). Groundwater levels have been monitored since the inception of the Ord irrigation project. Groundwater levels have continued to rise at alarming rates, of around 20cm per year over a significant proportion of the irrigated area. Due to the high water requirements of sugarcane and the significant area already under production or proposed for Ord Stage 2, sugarcane could potentially have a major impact on groundwater accessions. Concerns about an acceleration of rising groundwater levels associated with irrigated sugarcane production were a major reason for initiating this program of research. The sustainability of the Ord sugar industry will depend on the implementation of appropriate management practices which are compatible with both production and groundwater requirements. More recently, a groundwater management committee was formed by growers to develop and implement a land and water management plan for the area. The Ord Land and Water Management Plan was published in 2000 and includes goals for groundwater, irrigation efficiency and surface water quality. These include: 1. Reducing ground water levels to greater than 2 metres from the surface across the whole irrigation area within five years; 2. Holding the quality of groundwater at or above the quality in 2000; 3. Improving irrigation management to achieve 65% average annual water use efficiency on all irrigation farms within five years; 4. Reducing sediment loads in tail water by 40% within five years. Development of appropriate irrigation management practices for sugarcane will be a key requirement in the management of rising groundwater levels and in the sustainability of Ord production. Expansion of the Ord sugar industry into environmentally more sensitive areas adjacent to the existing irrigation area will also be dependent on the improvement of irrigation practices and a better understanding of water table management. 9 Climate Climatic conditions dictate a high water requirement for sugarcane growing in the Ord (Robertson et aI, 1997), and growers face a considerable challenge in meeting that requirement whilst maximising profitability and minimising deep drainage losses and potential impacts on groundwater and the environment. The unique climatic conditions and production system in the Ord irrigation area, with a long (30 week) harvest season, present new challenges to profitable and sustainable sugarcane production in Australia. With the establishment of a sugar industry in the Ord, important questions have been raised as to what is the potential sugar production per hectare given the climatic conditions in the region and what is the irrigation water requirement to reach this potential. Such information is essential not only for devising best practice irrigation management strategies for on-farm application but also for assessing the present and future allocation requirements for irrigation water. In an earlier SRDC project in the Ord, Muchow et al (1996) conducted a desktop study of potential annual sucrose yields and irrigation water requirements using the APSIM sugarcane crop simulation model for Kununurra, Ayr and Ingham. This indicative simulation analysis suggested average "potential" cane yields over a plant crop and four ratoons of 239 tonnes cane/ha for Kununurra, 201 tonnes cane/ha for Ayr and 191 tonnes cane/ha for Ingham. It also suggested an extremely high irrigation water requirement in the Ord which was more than double that in the Burdekin environment. The model predictions highlighted the need for field data on potential crop yields and crop water requirements under Ord conditions for sugarcane crops ratooned at different times of the year. Whilst irrigation scheduling guidelines have already been developed for different sugarcane areas in Queensland, it is unlikely that they can be directly applied to the Ord with its combination of very different climatic conditions (very high temperatures and evaporation) and a different suite of soil types. Allocatioll alld costillg ofirrigatioll water The current allocation of water by the Ord River Cooperative for surface irrigated crops is 17 ML/ha. Irrigation water is metered by means of a Dethridge wheel from the supply channels to a head ditch on each property. Current water costs for broad-acre crops are extremely low: $2.44/ML + $7S/ha/year fixed charge. Whilst at present only a small proportion of the available water resource in the Ord is being utilised, plans for new irrigation developments that will maximise the use of available water are now underway and this has introduced the need for a Water Allocation Plan. A draft Water Allocation Plan has been under review since 1999 and this will specifY the maximum amount of water that can be diverted per irrigation water year by the Ord Irrigation Cooperative. This amount will be a function of the current wet season rainfall and the procedures for its calculation specified in the water distribution operating strategy. The operating strategy will specifY water quality, water use and efficiency targets and will build on the community's commitments under the Ord Land and Water Management Plan. 10 2. PROJECT OBJECTIVES This project sought to benchmark current irrigation management practices in the Ord River Irrigation Area and to develop improvements to these that firstly maximised sucrose accumulation by defining appropriate "drying off' practices close to harvest; secondly that maximised profitability by matching irrigation applications with crop water requirements for different harvest times throughout tile April to November period; and thirdly that stabilised rising water tables and improved the efficiency of irrigation water use. The objectives of the project, that were set at the outset of the project, were as follows: 1. To benchmark current irrigation practices in the Ord Irrigation Area and hold an Industry workshop to define research priorities and the components of best practice irrigation management. 2. To conduct on-farm research to quantifY the impact of different irrigation practices on productivity, water use efficiency, profitability and groundwater management for the Ord sugar industry. 3. To collect yield accumulation and complementary climate, soil and crop data and use these to test a sugarcane crop growth simulation model tailored to Ord conditions. 4. To use field experimentation, databases and sugarcane simulation models to identifY best production practices related to irrigation management with the emphasis on maximising profitability by matching irrigation applications with crop schedules for different times of harvest. 5. To develop and promote irrigation management strategies that optimise the long-term profitability of the Ord Sugar Industry and keep accessions to the regional water table at a level required for sustainability. The project was reviewed by SRDC on 28-29 October 1997 at the Frank Wise Research Institute. The review team recommended that the project should continue into the expanded experimental phase envisaged at the start of the project. They made a number of recommendations regarding specific objectives of the project: • Support was given for the appointment of an Industry Development Officer to assist with tile development and promotion of best practice irrigation guidelines. • Best practice irrigation management needs to take into account soil variation across the irrigation area and local variations in clay content and soil depth can influence soil water extraction by the crop. • Initial experimental results from different experiments to quantifY water application efficiencies appear to be highly variable and care needs to be taken in the choice of suitable cane blocks for this work. • Although crop nutrition was outside the scope of the project, productivity variation may be related in part to nutritional limitations. Strategic monitoring of nutrient levels in both soils and plant tissues should be considered at experimental sites. • Project review and planning meetings should provide greater opportunity for dialogue and interaction. Consideration should also be given to coordinating the review and planning meetings to coincide with mid season and end of season sugar industry review meetings. • The formation of small focus groups of growers should be considered to facilitate the adoption, evaluation and demonstration of improved practices and the proposed Industry Development Officer should be encouraged to work closely with tllese groups. 11 3. PARTIES INVOLVED AND LINKS WITH OTHER INITIATIVES The project was set up as a research partnership between CSR Technical Field Department, CSIRO Tropical Agriculture (now CSIRO Sustainable Ecosystems), Agriculture Western Australia and SRDC. A key feature of the project was participatory on-farm research involving the Ord Sugar Industry, the research partners and SRDC. An important component of this participatory process has been the involvement of the Ord cane growers and Ord Industry Board in the planning and development of research activities. Research meetings were held twice a year throughout the course of the project to present results to Industry and progressively develop best management irrigation practices in collaboration with growers as appropriate information became available. This project when established was closely related to research being conducted by CSIRO Tropical Agriculture and CSR in the Herbert and Burdekin districts aimed at achieving efficient use of water and nitrogen in the sugarcane production system (SRDC projects CSC4S, 7S, 16S, 18S and 2IS). Immediate linkages were established between this project and the research being conducted in projects CSC 16S "Efficient use of water resources in sugar production: A physiological basis for crop response to water supply" (work mainly in the Burdekin district) and CSCI8S "Efficient use of water resources in sugar production: Optimising the use of limited water under supplementary irrigation" (work mainly in the Herbert district). It also had close linkages to a BSES project in the Burdekin focussed on improving water application efficiency - BS90S "Increased productivity through better design of irrigated cane fields". More recently, close links have been established with the irrigation risk management work being conducted in the Burdekin (project CTA038). The project was also closely linked to an Ord project initiated in 1995: WAAlS "Water table monitoring in the Ord River Valley" (Nulsen and Sherrard, 1999). This was a component of a larger initiative by Agriculture Western Australia and the WA Water and Rivers Commission to examine and implement appropriate options for the management of rising water tables in the Ord River Irrigation Area. This linkage was regarded as being critical to ensure that the options implemented to manage the rising water tables are consistent with the requirements for maximising Ord sugar industry profitability. The project also followed on logically from part of SRDC project ORD004S in which a desk-top analysis was conducted of the production potential and irrigation water requirements of sugarcane crops in the Ord using the APSIM sugarcane growth model and 80 years of historical weather data (Muchow et aI, 1996). 12 4. METHODOLOGY A partnership approach was adopted in this project to fund and service the needs of the Ord Sugar Industry. The partnership includes the Ord sugarcane growers, the Ord Sugar Industry, three research organisations (Agriculture Western Australia, CSIRO and CSR) and SRDC. An important part of this partnership approach has been the involvement of the Ord growers and Industry in the planning and development of research activities and in the interpretation and adoption of research results. The methodology adopted in this project involved the following sequence of activities: • Development of research partnership and preliminary consultation with growers and Industry representatives • Setting up automatic weather stations and collection of data on crop management, weather, soils and crop yields • Benchmarking of current irrigation practices using questionnaire surveys and crop logs of individual fields • Setting up instrumented experimental sites for the collection of data on yield and sugar accumulation, soil and climatic conditions, and on water use and water table levels • Conducting two sets of on-farm experiments which focus on different irrigation management strategies aimed at maximising sucrose accumulation by defining "drying off' practices close to harvest, maximising profitability by matching irrigation application with crop schedules for different harvest times, and stabilising rising water tables and improving the efficiency of irrigation water nse • Conducting crop simulation analyses using the APSIM-Sugar simulation model with data from the on-farm experiments to explore options for improved irrigation management • Reviewing research on irrigation water application efficiency and conducting experiments to develop ways of using irrigation water more efficiently • Appointing an Industry Development Officer to assist with the collection of experimental data on farms and to work closely with growers in developing more profitable and sustainable irrigation management practices. Further details of methodology are presented in tile sections of this report describing the individual components ofthe project (Sections 5 - II). 13 5. BENCHMARKING GROWERS' IRRIGATION PRACTICES Introduction A first element in addressing and ensuring best practice irrigation management in the Ord sugar industry was to obtain a full understanding of the range of current irrigation practices. To achieve this, a survey was undertaken to benchmark irrigation practices in the Ord Sugar Industry during the first year of commercial sugarcane harvest (1995-96 crop) in the Ord (Wood et al., 1998). The purpose of such a benchmarking activity was to identify those areas where potential improvements could be made and to guide decisions on research direction and investment to ensure that maximum benefit is delivered to the industry. It also provided a benchmark against which the impact of this research could be evaluated at a later date.' This section reports on the key findings of the survey and identifies options and research needs for further improvements in irrigation practice in the Ord sugar industry. Benchmark measures ofirrigation practice The following measures were used for benchmarking irrigation practices. These measures would apply equally well to other fully irrigated sugarcane areas in Australia: • Water applied per irrigation - length of run, siphon size, duration of irrigation and tail water discharge • Water applied per crop cycle - irrigation frequency at different times of year, length of drying-off, method of irrigation scheduling • Tonnes of cane and sugar produced per ML water applied • Tonnes of cane and sugar produced per ML water used by crops - irrigation application efficiency, rainfall effectiveness • Tonnes of cane and sugar produced per hectare, per annum • Cane sugar content (CCS, Pol) Methodology The survey included 13 growers and 27 blocks of sugarcane and was conducted by grower interviews during the 1995/1996 growing season. Yield data were available for 26 blocks, all of which were plant cane with cane yields between 117 and 210 tonnes cane/ha. These represented 438 ha and 79,823 tonnes of cane. Since the survey was condncted retrospectively, some of the variables are likely to be "best estimates" rather than absolute measurements. In particular, growers had very little scope to estimate irrigation application efficiency. The total water applied to each block and crop water use was calculated using the following methods: 1. Dates, times and flow rates in ML/hr were obtained from the Ord Irrigation Cooperative Limited for each block; 2. Growers supplied data on the number of hours water was applied per irrigation on each block; 3. Growers supplied data on the number of irrigations applied over the life of each crop, thus allowing the total amount of water applied in MLfha to be calculated for each crop; 4. The amount of water applied per annum was calculated using data supplied by growers on the age of the crop; 14 5. Crop water usage was calculated by mUltiplying the total water applied by the irrigation application efficiency estimated by the growers, and adding effective rainfall. The estimated application efficiency took into account the losses due to runoff down tail drains and deep drainage losses through the soil profile. It was assumed that 72% of total rainfall was effective in meeting crop water requirements (Muchow et ai, 1996). Of the 827 mm annual rainfall in 1995196, the amount of effective rainfall was calculated as 595 mm. Mill data for each block were used to calculate t cane per harvested ha. Since growers in the Ord are paid on Pol % cane, ccs for each block was derived using the relationship shown in Figure 5.1. This relationship was obtained using data from every rake of cane harvested in the ORIA in 1996 where both Pol and Brix were measured. Mill Pol and ccs data were adjusted for undetermined gains in the milling process using data derived from the CSR Summary and Season Forecast Report. Sugar yield was calculated as the product of cane yield and CCS/IOO, as in the Queensland sugar industry. 20r-------~------_.------__, CCS=1.045*Pol-1.736 r 20.97 n=16872 15 CI) <..> <..> 10 5~~----~------4-------~ 5 10 15 20 Pol Figure 5.1. Relationship between CCS and Pol on samples measured during 1996 in the ORIA. The mean aud range of the different variables were calculated, and compared with the block that achieved the highest sugar yield (e.g. Table I). The survey results were compared with the results of preliminary model predictions of crop water requirements. Results and Discussiou Analysis of the survey results indicates that a wide range of irrigation practices are employed by sugarcane growers in the Ord. Table 5.1 summarises key water management data which have been collected, estimated or calculated from the survey. The data support the results of preliminary analyses using the APSIM sugarcane model which suggested that sugarcane crops in the Ord have a much higher irrigation water requirement than those in other sugarcane districts (Muchow and Keating, 1998). A huge range in the number of irrigations, 15 amount of water applied and length of drying off period was recorded. This suggests that growers are using a range of water management strategies to achieve profitable sugarcane production and would benefit from assistance in developing the most appropriate strategy to achieve the most efficient and profitable use of irrigation water. The total water applied to each crop is extremely high ranging from 20 to over 80 ML ha-' . The amount applied to the block with the highest sugar yield is about the same as the mean. Estimates of irrigation water application efficiency were supplied by growers for 15 of the 27 blocks surveyed and these ranged from 60 to 75%. Comparisons with other sugarcane producing districts are difficult as very little benchmark irrigation data are available from other areas apart from recent work on irrigation water application efficiency in the Burdekin district (Shannon and Raine, 1996; Raine and Bakker, 1996). In these studies, irrigation application efficiency varies with soil type and length of irrigation run, with efficiencies on low infiltration cracking clay soils in excess of 70%. In an earlier study with kenaf in the ORIA, Muchow and Wood (1981) measured application efficiencies ranging from 50 to 85% on Cununurra clay, with application efficiency decreasing with an increase in irrigation frequency. Fortnightly irrigation schedules resulted in application efficiencies close to 70%. In that study, significant infiltration of water into the soil still occurred when lateral wetting of the soil surface was complete and the flow of water from channel to field was stopped. Water application efficiency with flood irrigation on Cununurra clay for maize production ranged from 67 to 86% (Sherrard et al., 1991). There is a clear need to obtain better estimates of both runoff and deep drainage by conducting field measurements in the ORIA. Table 5.1. Total water applied during irrigation. Measure Mean Range Value for block with highest sugar yield Total number of irrigations Length of run (m) Total water applied (ML ha-' crop-I) Irrigation application efficiency (%)# Total crop water use (ML ha-' crop-')** Crop water use efficiency (t cane ML-')** Drying-off before harvest (d) Annual water applied (ML ha-' yr"') Annual crop water use (ML ha-' yr"')** 27 (27/ 489 (27) 40.6 (26) 70 (15) 31.9 (15) 6.1 (15) 36 (27) 32.5 (27) 27.5 (15) 15 -42 250-1100 20.8 - 83.1 60 -75 24.2 - 45.3 3.9 - 7.6 7 - 82 15.3 -53.8 22.7 - 41.0 37 1100 39.8 65 31.8 6.6 42 34.1 28.1 + Number of blocks with available data indicated in brackets; # Grower estimate only; **Assumes effective annual rainfall of 595mm and estimated application efficiency accounts for runoff and drainage losses. Crop water use was calculated by assuming that 72% of total rainfall was effective in meeting crop water requirements (Robertson et al., 1997) and using the estimated application efficiency. The average crop water use per annum was 27.5 ML ha-', with values ranging from 22.7 to 41 ML ha-'. Preliminary simulation analyses have suggested a crop water requirement of 29.3ML for a 212 t cane ha-' twelve month sugarcane crop growing under 16 ideal conditions in the ORIA (Muchow et al., 1996). Whilst the crop simulations have not been validated for ORIA conditions, further model runs (Robertsou et al., 1997; Muchow and Keating, 1998) suggest that these estimates are realistic and the maximum likely. Consequently, it is unlikely that the high value of 41 ML ha·l , at the upper end ofthe range of estimated annual crop water use, is correct. This is most likely to be due to overestimates of application efficiency. This further highlights the need to obtain better estimates of application efficiency. Calculated values for crop water use efficiency for cane crops harvested in the 1996 season (mean 6.1 t cane ML·1, and 6.6 t cane ML·1 for the block with the highest sugar yield) are similar to the simulated crop water use efficiency using 81 years of climate data of 7.24 t cane ML ·1 for a 12 month ratoon crop grown under non-limiting conditions (Muchow et al., 1996). From the survey data, there was no clear relationship between sugar yield and the amount of irrigation water applied (Figure 5.2). With such variation, it is not possible to use the survey data to identify "best-bet" strategies for maximising sugar yield in relation to irrigation application. Rather this survey has highlighted the need for field experimentation coupled to simulation modelling to explore the relationship between sugar yield and water applied in order to identify best-practice irrigation management. 30 .....-. 25 I- ---..IcII: -C iii 20 - 's..;., III C"l -::J CI) 15 1I I . .. ., • .*•.*. . • • . * • • .. ..• * I - - • ... *Q96 • Q95 NCo310 Q99 10 I I I I 0 20 40 60 80 100 Total water applied (ML ha -1 crop-1) Figure 5.2. Variation in sugar yield for four varieties with different amounts of irrigation water Data on irrigation frequency and the number of irrigations (Table 5.2) show that the climatic conditions of the Ord River area demand very frequent irrigations, particularly in the hotter parts of the year. Whilst mean data for the frequency of irrigation did not change much over the year, irrigation frequency on the highest yielding block increased to weekly irrigations for the August to November period. 17 Table 5.2. Frequency of irrigation and number of irrigations. Months Frequency of irrigation (days) Number of irrigations Mean Range Best Block Mean Range Best Block Apr. 95 - Jul. 95 15 (22t 10-21 14 5 (22t 2-9 4 Aug. 95 - Nov. 95 14 (26) 7-21 7 9 (26) 6-15 15 Dec. 95 - Mar. 96 13 (24) 7-20 7 6 (27) 4-9 9 Apr. 96 - Jul. 96 Aug. 96 - Nov. 96 13 (27) 14 (13) 7-20 0-30 10 7 (27) 3-15 - 4 (12) 0-10 9 - + Number of blocks with available data indicated in brackets Considerable wastage of irrigation water is suggested by the data on tail drain flow in Table 5.3. The need for such frequent irrigations and the costs and feasibility of applying irrigation water in this way require further investigation. Furthermore the need to maximise profitability rather than productivity and the need to minimise accessions to the ground water need to be considered in developing best-practice irrigation management. Table 5.3. Duration of irrigation and tail drain flow. Months Duration of irrigation (hours) Mean Range Apr. 95 - Jul. 95 20 (22t 12-24 Aug. 95 - Nov. 95 20 (26) 12-24 Dec. 95 - Mar. 96 20 (27) 12-24 Apr. 96 - Jul. 96 20 (27) 12-24 Aug. 96 - Nov. 96 20 (12) 12-24 + Number of blocks with available data indicated in brackets Time to reach tail drain (hours) Mean Range 14(18t 3-20 15 (22) 3-20 15 (23) 5-20 15 (22) 5-20 15 (11) 7-20 Duration of tail drain flow (hours) Mean Range 7(2W 3-10 7 (24) 3-17 7 (25) 3-17 7 (24) 3-17 8 (10) 4-17 The very large variation in the amount of water applied to sugarcane crops in the Ord is confirmed by the data in Table 5.4. The amount applied to the highest yielding block is well below mean applications. Also, the average amount of water applied per irrigation is surprisingly high for the April to November period when crops would have been relatively small. This suggests that application efficiencies may have been considerably lower than the 60-75% estimated by growers (Table 5.1). 18 Table 5.4. Water applied. Months Water applied per irrigation (MLha-') Mean Range Best Block Apr. 95 - Jul. 95 3.8 (22t 0.4-15.2 1.1 Aug. 95 - Nov. 95 3.3 (26) 0.3-15.2 1.1 Dec. 95 - Mar. 96 3.2 (27) 0.3-15.2 1.1 Apr. 96 - Jul. 96 Aug. 96 - Nov. 96 3.1 (27) 1.8 (12) 0.3-15.2 0A-8.4 1.1 - Total water applied (MLha-') Mean Range Best Block 18.7 (22t 1.6-106.3 4.3 26.0 (26) 1.8-121.5 16.1 15.9 (27) 2.3-75.9 9.7 17.4(27) 6.5 (11) 1.5-62.8 1.1-33.6 9.7 - + Number of blocks with available data indicated in brackets Cane and sugar yields in the Ord were extremely high during the 1996 harvest season compared with cane growing areas in Queensland (Table 5.5). Since the Ord sugar industry is in its early developmental stages, all of the blocks monitored for this survey were plant cane with a comparatively long crop duration (mean crop age 447 days). There was no clear relationship between sugar yield and harvest date or crop age. Crops of greater age did not necessarily produce higher sugar yields. A yield plateauing phenomenon has been observed in high yielding, long duration crops in Queensland (Muchow et aI, 1995). Such a phenomenon may also be occurring in these high yielding Ord crops. Growth analysis studies to quantify yield accumulation over time are therefore required to assess the best crop age for harvest. They will also help explain why current crops appear not to be reaching their yield potential, as predicted using the APSlM model (estimated sugar yields 30.9 tonnes sugar/hectare for a 12 month ratoon crop, Muchow et at., 1996). Table 5.5. Productivity measures. Results from a survey of 12 growers and 26 blocks. Measure Mean Range Value for block with highest sugar yield Cane yield (t ha-') Pol % cane * CCS * *Sugar yield (t ha-J ) Crop age (d) Block area (ha) Harvest date 183 117-210 209 13.6 10.8 - 16.0 14.3 12.4 9.4 -15.0 13.2 22.9 16.8 - 27.5 27.5 447 305 - 576 443 16.8 2.5 - 45.0 36.3 - 13/06/96 - 10/12/96 25/07/96 '" Data only available from 11 growers and 24 blocks 19 Values of pol % cane and ccs from the 1996 harvest season were lower than anticipated and this has proved to be a m'\ior issue affecting the profitability of the Ord Sugar Industry. Whilst drying off before harvest is used to maximise sucrose content, the survey data show no clear relationship between ccs and duration of drying off (Figure 5.3). These data are confounded with different climatic conditions during the period of drying off before the different harvest dates. Since drying-off is an important management strategy available for sugarcane growers in the Ord, there is a need to collect data on crop performance with drying off at different times during the harvest season, so that the findings on drying-off from Queensland projects can be compared with the Ord. 16r----r----r----r---~1r---~ 14 '- (J) o 12 C- O 10 I- •• • ... • • * ... * * • ... • •* •• * ... • ... • Q96 • Q95 ... NCo310 - • - - * Q99 8~---L-1---~1--~1-----'~--~ o 20 40 60 80 100 Drying-off (d) Figure 5.3. Variation in ccs with different durations of drying off before harvest. Conclusions The benchmarking survey of irrigation practices confirmed that "best practice" irrigation management had not been developed for sugarcane crops and that many of the growers were applying large amounts of irrigation water at frequent and reasonably uniform intervals throughout the year. The results indicated a number of key information needs: 1. Better estimates of application efficiency and components of the soil water balance. 2. Better definition of water requirements in relation to crop growth and appropriate irrigation schedules that maximise profitability for different crop start and harvest dates. 3. Climate-Crop-Soil-Management data from field experiments and crop logs that can be used to validate simulation models for testing hypotheses and developing options for best practice irrigation. 4. Better information on crop management and the drying-off requirement to maximise sugar content (pol, ccs) and sugar yield for different harvest times throughout the year. 20 6. DRYING OFF EXPERIMENTS Introduction In irrigated sugarcane production, water is commonly withheld prior to harvest in order to dry out the field and make it suitable for mechanical harvesting operations. In the Burdekin area this practice also increases both the sucrose concentration of the cane as well as the sucrose yield (Robertson et aI., 1999a; Robertson et ai., 1999b). The optimum length of the drying off period depends on the rate of development of the crop water deficit and the associated changes in sucrose yield, cane yield and sucrose concentration. The imposition of a water deficit through drying off commonly reduces stalk yield through reduced carbon assimilation but leads to an increase in sucrose concentration through increased partitioning of dry matter within the plant towards sucrose. Generally drying off is more often associated with an increase in sucrose concentration rather than with a reduction in cane yield (Robertson and Donaldson, 1998). In the early irrigation experiments with sugarcane in the Ord, little or no response in sucrose yield was obtained to drying off (Kingston et aI., 1980; Gallagher, 1986). It was concluded that a 3-5 week drying off period was adequate for most of the harvest season and that if the drying off period exceeded 6-8 weeks then the cane tends to deteriorate with losses of both cane yield and sucrose concentration (Bureau of Sugar Experiment Stations, 1982). It was decided to test these guidelines by conducting a series of field trials to quantity the impact of different drying off strategies on cane yield and sucrose concentration. Drying offexperimelZts Seven field drying off experiments were conducted during the period 1996-1998 (Table 6.1). Table 6.1. Drying off experiments conducted in the Ord. BlockNol Plot size 230.3.10 1.24 ha Varietyl Crop Class Q991R 300.11.20 Q991R 1.81 ha 220.21.10 Q951R 2.43 ha 230.3.10 1.24 ha Q992R 300.11.20 Q992R 1.81 ha 220.21.10 Q952R 2.43 ha FWI3B 0.778 ha Q99P Crop Start Date 02/05196 01/08/96 20109/96 18/09/97 17/08/97 29/08/97 13/05197 Harvest Date 18/09/97 17/08/97 29108/97 29/10198 22/07/98 2110/98 7107198 Drying-off Treatment (days) 42 n 82 48 62 74 20 30 43 20 37 37 38 52 66 40 54 68 24 47 80 Cane Yield (thaI) 143 138 144 147 143 145 131 127 135 159 147 154 149 155 148 147 146 143 193 184 180 Adjusted Pol Yield Pol (t ha-I) 13.6 19.3 13.6 18.7 13.6 19.5 14.5 21.3 14.2 20.3 14.2 20.6 15.5 20.4 16.4 20.9 15.9 21.6 11.4 18.2 12.0 16.9 12.2 18.8 14.0 20.8 13.4 20.8 13.4 19.9 15.2 22.6 15.1 22.0 15.7 22.4 10.6 21.1 10.6 19.5 10.5 19.0 21 Six of the experiments were located on commercial fields with the crops being grown using the commercial practices of individual growers. The other experiment was located on Block 3B at the Frank Wise Institute Research Station. Blocks were selected with a minimum of 200 drills to allow the different drying off treatments to be spaced out so that lateral soakage would not be a problem between treatments. Each treatment occupied at least 10 drills so that the rakes would be weighed and analysed separately at the mill and each drying off treatment was replicated three times. The response to drying off prior to harvest was examined by withholding irrigations for varying times prior to harvest. The drying off treatments were chosen by the growers in consultation with the research team. Each grower was responsible for applying the treatments and for completing a crop log which detailed the crop management activities on the particular trial block. The cane in each replicate was mechanically harvested and both cane yield and Pol were determined at the Ord sugar mill. In addition, for the four experiments conducted in 1997-1998, treatments were sampled by hand at least twice before the drying off treatments were imposed and again at final harvest. Total fresh weight, millable stalk fresh weight and dry matter content were recorded at each sampling and a stalk count was conducted. Also juice samples were obtained after fibrating and processing the samples through a Carver Press for determination of brix, pol and fibre. No rainfall was recorded during the drying off periods. Results 1996-97 experiments In the 3 experiments conducted during 1996-97, the drying off periods ranged from 3-7 weeks, 4-10 weeks and 6-12 weeks. Results of cane yield, adjusted Pol and Pol yield by commercial harvesting are given in Table 6.1. Statistical analysis showed that there were no significant effects of drying off on either cane yield, Pol or Pol yield, which is surprising given the extended drying off periods of up to 82 days. Both cane yields and adjusted Pol readings were remarkably consistent across treatments. 1997-98 experiments The experiments were repeated on the same blocks as in 1996-97, although the drying off periods were different. Again the cane yield and Pol readings obtained from the 3 commercial farms were very similar between treatment and showed no significant differences. An additional drying off trial was conducted in plant cane in Block 3B at the Frank Wise Institute research farm. The results in this trial did show reduced cane yields with increasing drying off periods, although there was no difference in adjusted Pol between treatments, even though the cane harvested from the 80 day drying off period appeared severely dessicated with virtually no green leaves. Results of the hand-samplings conducted on the four trials in 1997-98 are given in Figures 6.1-6.4. In only one trial, Block 3B at the Frank Wise Research Institute, were there any significant differences between treatments (Table 6.2). In this trial, the 80 day drying off treatment had a significantly lower CCS, Pol % cane, Pol yield, purity, and on a dry weight basis a significantly lower stalk sucrose concentration and significantly higher stalk fibre concentration and stalk nitrogen concentration. Most of these, however, were not significant for the 47 day treatment. 22 Table 6.2. Results of drying off experiment conducted at Block 3B, Frank Wise Institute. Sample 6 (6/7/98), 419 days after planting. (* P<0.05, *. P, c, e""' 10 ="> (J) 5 0 200 250 300 350 400 Days after planting 0.15 §" 440 Dry-off ,lL 0.12 • 80 days Cl • 47 days .2! c: • 24 days .2 0.09 "§ E (J) 0.0 450 200 250 300 350 400 Days after planting 450 Figure 6.1. Effect of three drying off treatments on cane yield (t ha·I), cane biomass (t ha·I), sucrose yield (t ha·l) and fresh weight sucrose concentration (g g"1) in the drying off trial conducted in Block 3B at the Frank Wise Research Institute (trial # 440). (Sucrose yield aud sucrose concentration were significantly different (p ::> (f) 5 0 180 240 300 360 Days after ratooning 0.20r----.-----,-----,----, ~ 442 Dry-off LL • 37a days '0> 0.15 .. 37b days .9 .. 20 days oc: '"i"E" 0.10 OJ c<.:> 8 g~ 0.05 ::> (f) 0.0 '-----'------'---"------' 420 180 240 300 360 420 Days after ratooning Figure 6.3. Effect of three drying off treatments on cane yield (t ha-I ), cane biomass (t ha-I ), sucrose yield (t ha-I) and fresh weight sucrose concentration (g il) in the drying off trial conducted in Block 230.3.1 0 at Pegg's farm (trial # 442). 26 443 Dry-off • 66 days .. 52 days "38d~ ~"0 100 cQ) <3 50 50,--.--,---.--.---.--, 443'D'Y-Off • , , "",~ 40 .66d~YS .. 52 days .c .. 38 day' t::.. 30 'E''""" ~ 20 cQ) ('") 10- _ - - O~-J--~--~--~--~~ 200 225 250 275 300 325 350 Days after ratooning o '" , 200 225 250 275 300 325 350 Days after ratooning 25 443 Dry-off 20 '", .c 15 "0 ~ §Q) 10 => (J) 5 .. 66 days .. 52 days .. 38 days __ __O~-J ~ - L_ _~_ _L-~ 200 225 250 275 300 325 350 Days after ratooning c .o~ 0.09 E 1l c 0.06 8 443 Dry-off gQ) 0.03 .. 66 days .. 52 days => (J) • 38 days 0'£00 225 250 275 300 325 350 Days after ratooning Figure 6.4. Effect of three drying off treatments on cane yield (t ha-'), cane biomass (t ha-'), sucrose yield (t ha-') and fresh weight sucrose concentration (g i') in the drying off trial conducted in Block 300.11.20 at Cummings Brothers' farm (trial # 443). Discussion Robertson and Donaldson (1998) reviewed many drying off trials conducted in South Africa and Australia. They showed that the likelihood of a response to drying off is highly variable. The absence of a response to drying off in terms of sucrose concentration or stalk dry matter content in these trials could be associated with sugarcane crops in the Ord being capable of extracting large amounts of water from the soil profile. It may also be related to low rates of crop water use. The data presented in Figures 6.1-6.4 suggest that cane growth, and consequeutly crop water use, was generally quite slow when the drying off treatments were imposed. Consequently the drying off treatments were not able to impose much stress on the growth of the crop or initiate an increase in SUcrose accumulation. This issue will be further discussed in the next chapter. 27 It therefore appears likely that drying off cannot be used as a strategy for increasing cane sucrose content in the Ord unless the cane is stilI growing rapidly when drying off commences. Where cane growth has already slowed down, the main advantage of drying off will be to dry the soil surface, induce cracking in the soil and improve trafficability. Given the findings of these 7 trials, it is recommended that the drying off strategy should be tailored to allow fields to be dried out sufficiently for mechanical harvesting. Extending the drying off period would promote cracking and would benefit soil structure (P. McCosker, pers. comm.) and would also result in labour and water savings associated with less irrigation. However the results from the experiment at Block 3B at the Frank Wise Research Institute indicate that it is possible to dry off for too long. The 80 day dry down treatment resulted in a significant reduction in CCS and purity. 28 7. IRRIGATION SCHEDULING TRIALS Introduction In section 1 it was noted that much of the early irrigation scheduling work with sugarcane in the ORIA was inconclusive with little or no response to irrigation frequency. Consequently when commercial sugar production began in the Ord, irrigation schedules of 25-30 irrigations per crop year were adopted. Watering the crop every 10-14 days throughout the year was regarded as being necessary because sugarcane crops growing in the unique climatic conditions of the Ord had an extremely high water requirement. To quantify the yield responses to different irrigation schedules a field experimental programme was conducted so that improved irrigation practices could be developed which not only maximised sugar yields and profitability but also improved the efficiency of irrigation water use and led to a stabilisation of rising water tables. Choice oftreatments The irrigation scheduling trial was established in 1998. Treatments imposed at canopy closure in the first year were 60, 120 and 180 mm of cumulative daily evaporation (Class A Pan). Treatments were imposed throughout the season until dry down, five weeks prior to harvest in July 1999. After analysis of the data and consideration of the results it was decided that the treatments in the second year of the trial should be based on soil moisture deficit. The quantity of water available to sugarcane over a depth of 1.9 metres was calculated to be 225mm. Irrigation was scheduled after the sugarcane had removed of 79, 123 and 191 mm of the available water. Irrigation prior to canopy closure was scheduled at 120mm cumulative evaporation for both crops. Growth analysis data PZant crap experiment Stalk popUlation declined in all treatments up to about 300 days after planting and then remained at about 8 stalks per m2 (Figure 7.1). Stalk popUlation often decreases slowly at this stage due to lodging or water stress but there was no sign of these stresses in this experiment in regard to stalk popUlation at least. 1 ~ '"E 1~ ~ ..'a" E c'" ]-'i" (f) .....-m..-.._.._.._.....-.. ........• Irrig 60mm pan evap • Irrig 120mm pan evap " Irrig 180mm pan evap 100 200 300 400 Days after planting 500 Figure 7.1. Stalk population for three irrigation scheduling treatments applied to a plant crop Q99 in the Ord (199811999) 29 Green leaf area index (LA!) was as high as 6.0 in the most frequent irrigation regime (60 mm pan evaporation). LA! decreased over time (Figure 7.2) due to declining stalk population and declining leaf area per stalk (Figure 7.3). LAI stabilized after 300 days (Fig 7.2). The only indication that the water was limiting in the 180 mm treatment was the reduced LAI in this treatment at the time of the second sampling. •••••'. .. .. Irrig 60mm pan evap Irrig 120mm pan evap ........ II Irrig 180mm pan evap 100 200 300 400 Days after planting 500 Figure 7.2. Green leaf area index for three irrigation scheduling treatments applied to a plant crop Q99 in the Ord (199811999) 800'(\-----,-----,----,------, 7i -2. 600 -'" :m -m 400 ~'" 200 •.... ... -- - .... ... - -...-..-...........'t.:..................- . .. .. Irrig 60mm pan evap Irrig 120mm pan evap ........ II Irrig 180mm pan evap 100 200 300 400 500 Days after planting Figure 7.3. Leaf area per stalk for three irrigation scheduling treatments applied to a plant crop Q99 in the Ord (199811999) Cane biomass (or stalk dry matter yield) increased rapidly in all treatments but was reduced significantly by lack of irrigation at the time of the 2"d and 4th samplings (Figure 7.4, Table 7.1). Had the crop been harvested at 12 months cane biomass yield would have been reduced significantly by allowing 180 mm pan evaporation to accumulate rather than irrigating more frequently at 120 or 60 mm deficit. However the crop was harvested at 15 months even though there was little further yield accumulation in cane stalks. At harvest there was no difference between treatments in regard to cane biomass yield. 30 6 ,~ ~ .rt=il e. (J) (J) til E 0 13 E~ r:- "0 -]Y'" (fJ o. o. o. 0.1 O. 100 ...."- " ...... _.!._Jf.J' , r ...·· III Irrig 60mm pan evap ........III Irrig 120mm pan evap - - _.III Irrig 180mm pan evap 200 300 400 Days after planting 500 Figure 7.S. Stalk dry matter content for three irrigation scheduling treatments applied to a plant crop Q99 in the Ord (1998/1999) 31 22 ,~ ~ rn 16 .<:: ~ 1a:;l 11 ';;' Q) crn: () 5 100 • Irrig 60mm pan evap • Irrig 120mm pan evap II Irrig 180mm pan evap 200 300 400 Days after planting 500 Figure 7.6. Cane yield for three irrigation scheduling treatments applied to a plant crop Q99 in the Ord (1998/1999) Sucrose concentration on a dry (Figure 7.7) and a fresh weight basis (Figure 7.8) decreased significantly with increased irrigation prior to the wet season. The crops in the 120 and 180 mm irrigation regimes were therefore under some degree of water stress resulting in a 'ripening' response associated with greater partitioning of photosynthate to sucrose in the stalk than in the 60 mm irrigation regime. In April (349 days) differences in the opposite direction were significant for sucrose content of dry mass (Table 7.1). This demonstrates the dynamic nature of the effect of irrigation regime on sucrose content which can increase initially due to changes in carbohydrate partitioning but can then decrease in stressed plants because of retarded development. Growth retardation will result in shorter stalks with a greater proportion of immature top than in better developed stalks. Although sucrose content at harvest was not significantly different between treatments (Table 1) it is reasonable to assume that the increase in sucrose content was in fact due to increased irrigation because of the significance of an earlier result and also because of the significant effect of treatment on sucrose yield.. ~ 0 O. ,~ Ol -'?l O. gc: ~ O. C (]) : O. 0 <.> (]) ~ 0.1 0 <.> '"(J) O. 100 ..•..•-••- :...r .A .11·"_ _ .... - ...............'" •• .......... Irrig 60mm pan evap .. Irrig 120mm pan evap II Irrig 180mm pan evap 200 300 400 Days after planting 500 Figure 7.7. Sucrose content on a dry weight basis for three irrigation scheduling treatments applied to a plant crop Q99 in the Ord (199811999) 32 0.2Q-----,-----,-----,------, ,~ • Irrig 60mm pan evap Ol • Irrig 120mm pan evap •••••••• .9 oc: 0.1 II Irrig 180mm pan evap - - - • "g" c: ID 0.1 Uc: 8 ID -.......:,•.'•......-....-..... § :0 UJ 0 . 0 - - - - . . 1 - - - - - ' - - - - - ' - - - - - - ' 100 200 300 400 500 Days after planting Figure 7.S. Sucrose content on a fresh weight basis for three irrigation scheduling treatments applied to a plant crop Q99 in the Ord (1998/1999) 100 200 300 400 500 Days after planting Figure 7.9. Sucrose yield for three irrigation scheduling treatments applied to a plant crop Q99 in the Ord (199811999) Sucrose yield increased significantly with irrigation in April and July (Figure 7.9, Table 7.1). A similar effect was obtained from the yields determined by mechanically harvesting the experimental plots, although the results were not significant (Table 7.2). Given the low cost of water, the most frequent irrigation treatment would probably be justified on the basis of these results. Cost of irrigation is estimated to be about $3 per irrigation per ha. The 60 mm schedule received 9 more irrigations than the 120 mm schedule costing about $271ha more. A sucrose yield benefit of 1.6 tlha in April and 1.9 tlha in July by following the 60 mm rather than the 120 mm pan evaporation schedule would probably be justified when considering on farm costs. However one may ask if the additional yield is really worth the large amount of additional water considering the likely adverse effect on the water table and on the quality of water running offthe farm and hack into the river system. 33 Table 7.1. Plant crop results of irrigation scheduling trial (#444) growth analysis, Block 2B, Frank Wise Institute, Q99 Planted 06/05/98. (* p<0.05; •• p/+.+ . , ........I' "- ... o 100 200 300 400 500 Days after ratooning Figure 7.11. Green leaf area index for three irrigation scheduling treatments applied to a first ratoon crop Q99 in the Ord (1999/2000) Cane biomass increased rapidly up to 245 days after ratooning and then more gradually with very little growth observed between 337 and 406 days after ratooning. Cane biomass accumulation in the 191 mm treatment practically ceased after 245 days (Figure 7.12) but treatment differences were not significantly different at any stage (Table 3). As was the case in the plant crop, cane biomass accumulation was associated with increasing dry matter content (Figure 7.13) more than with an increased cane yield (Figure 7.14). Yields exceeding 36 100 t canelha were measured after only 245 days of growth but more than 90% of this material was water. Mean cane yield increased only to 133 tlha, 337 days after ratooning. Irrigation treatment had no significant effect on cane yield. The lack of response to irrigation and the low yield raises the question about factors other than water that may have been limiting yield. .p- o ...... .'<"::4 ;t:. g"J' E ~2 Q) c '"() o Irrig 79mm swd Irrig 123mm swd Irrig 191 mm swd ........ ..-.-.. ..""-.'....:.. .....- -••.•..•. •< 100 200 300 400 Days after ratooning 500 Figure 7.12. Cane biomass for three irrigation scheduling treatments applied to a first ratoon crop Q99 in the Ord (1999/2000) ~ 0.4 ,..---,.---..,....----r---,----, ~ : Irrig 79mm swd Irrig 123mm swd o Ol 0.3 .. Irrig 191 mm swd ~ Ol ';:" 0.2 ~ E C:- 0.1 u "ro'" 1i5 0.0 0 100 200 300 400 500 Days after ratooning Figure 7.13. Stalk dry matter content for three irrigation scheduling treatments applied to a first ratoon crop Q99 in the Ord (1999/2000) 37 22r---~~.;-~~~:s~-=~=-~ Irrig 79mm swd • Irrig 123mm swd ........ f'16 .. Irrig 191 mm swd --.,_.- .. . -....'".<:: ..... :!::C "0 11 (jj J "';;' til .l- offi 5 /, a 100 200 300 400 500 Days after ratooning Figure 7.14. Cane yield for three irrigation scheduling treatments applied to a first ratoon crop Q99 in the Ord (1999/2000) Sucrose content on both a dry and fresh weight basis, reached slightly higher values than in the plant crop, but sucrose content in this experiment was nowhere near the maximum values often obtained in Queensland (0.55 and 0.18 on dry and fresh mass bases respectively). Sucrose content for the 191 mm treatment decreased between 337 and 406 days after ratooning (Figures 7.15 and 7.16) and the treatment effect on both expressions of sucrose content was significant at harvest. Juice purity in this treatment also decreased before harvest probably because water stress was severe enough to lead to senescence in some stalks (Table 7.3). 0.6 0> 0.5 -S r:: 2 0.4 - ~ r:: 0.3 - ....2l ruo:: 0.2 - Irrig 79mm swd til ~ 8 0.1 - Irrig 123mm swd II Irrig 191 mm swd ::J C/) 0.0 0 100 200 300 400 Days after ratooning 500 Figure 7.15. Sucrose content on a dry matter basis for three irrigation scheduling treatments applied to a first ratoon crop Q99 in the Ord (1999/2000) Sucrose yield in July (337 days) and September (406 days) was reduced significantly by the less frequent irrigations compared to frequent irrigations applied when soil water deficit was 79 mm (Figure 7.17). There was a penalty of 1.3 t sucrose /ha in July and 1.6 t sucrose/ha in September by scheduling irrigation at a 123 m deficit rather than a 79 mm deficit. However eight irrigations were saved whilst incurring this small yield loss. This result is similar to that obtained in the plant crop experiment where it was concluded that the irrigations saved may not be economically justified by a sucrose yield penalty of about 1.5 tlha. Sucrose yield loss in the 191 mm treatment was 3.0 and 5.3 tlha respectively in July and September. Irrigating at this large deficit saved II irrigations, three more than the 123 mm treatment but the yield loss was disproportionate to the water saved. The choice of the 123 mm deficit irrigation schedule would be a good compromise between water savings and yield loss. 38 ........,~ 0.20 n••--':'r::;ri::-gi=;7;;-9m::::::m~s::w:JI---:::::I==--'--' • Irrig 123mm swd OJ ~ • 0.15 Irrig 191 mm swd o0::: i 0.10 g'" ~8 '" 0.05 if) 100 200 300 400 500 Days after rataaning Figure 7.16. Sucrose content on a fresh weight basis for three irrigation scheduling treatments applied to a first rataon crop Q99 in the Ord (1999/2000) 3 • Irrig 79mm swd • Irrig 123mm swd • Irrig 191 mm swd 2 ........ - - -- o 100 200 300 400 500 Days after rataaning Figure 7.17. Sucrose yield for three irrigation scheduliug treatments applied to a first ratoon crop Q99 in the Ord (199912000) Yields obtained from mechanically harvesting these plots were slightly different to the yields obtained by hand sampling (Table 7.4). The 123 mm treatment gave the highest commercial cane and sucrose yield (Table 7.4) and the 191 mm treatment gave the lowest yields. However none ofthese differences were statistically significant. 39 Table 7.3. First ratoon results of irrigation scheduling trial (#445) growth analysis, Block 2B, Frank Wise Institute, Q99 IR Ratooned 04/08/99 * p<0.05; •• p--.. l\ "'- / --..........:'-. I ~/ I/ ~ f~ If .! 0"-",, / '/ \""" \. --....,,;/ ~ './\./ 440 ~ "'>;A. T l"r"', I J-4.". I \~ I \. "l! \ 320 300 =o§ 280 .~s "260 ~ e 240 ~ o 220 ! '0 200 "' Figure 7.19. Daily cane stalk elongation and soil moisture to a depth of 800mm. The results from B27 also indicate that as soil moisture depleted stalk elongation decreased. However, these results also indicate some inconsistency in the relationship between soil moisture levels and rate of elongation with different rates occurring at different soil moisture levels for different irrigation cycles. Other blocks examined utilised the BSES method of irrigation scheduling ie. re·irrigating once rate of elongation reduces to 50% of maximum. 47 For a crop at canopy closure in September - December, this method indicated that reirrigation should occur approximately every 7 days. This result was inconsistent with irrigation trial results which indicated that less frequent irrigation would result in similar yield at harvest, possibly resulting from compensatory 'catch up' growth for less frequently watered crops, with wet season rains. Based on the above results, further investigation would be required before utilising this method commercially in the Ord. 48 8. WATER EXTRACTION BY SUGARCANE ON DIFFERENT SOILS Introduction In irrigated cropping systems the goal is to mID(lmlse profitability whilst exploiting opportunities to save water and improve crop water use efficiency and hence conserve the environment. Determining the best-bet irrigation strategy requires knowledge of the crop water requirement at different stages of growth under variable climatic conditions and the capacity of the soil, both to store rainfall and irrigation water and to supply water to meet crop growth requirements. The soil water holding capacity depends both on the available water capacity in the soil and the depth of crop extraction. Sugarcane is a new commercial crop in the ORIA and little information is available on plant extractable soil water (PESW) and soil water extraction at different depths of the soil profile for sugarcane growing in the ORIA. This information is required to parameterise cropping systems models such as APSIM Sugarcane (Keating et 01., 1999). Given year-to-year variability in climate, different soil types and variable cropping cycles in sugarcane production, cropping systems models are a valuable, if not essential tool, to determine irrigation water requirements and yield responses to different irrigation schedules. The use of the modelling approach with historical climatic data overcomes the year-specific and sitespecific limitation of field experimentation. There is limited information on maximum extractable water per unit soil depth for sugarcane in Australia (Inman-Bamber et 01.2000). Baran et 01. (1974) found 180 mm was extracted from a 2.0 m profile by sugarcane and Koehler et 01. (1982) recorded extractable soil water of 160 mm to 1.5 m. For a grey cracking silty clay, Inman-Bamber et 01. (1999) observed that water was extracted at least to a depth of 1.8 m and the PESW was 130 mm, with considerable extraction below 1.0 m. Thompson (1976) found that irrigated sugarcane did not remove water from depths greater than 90 cm on a clay soil, but water was extracted to a depth of 180 cm on a sandy soil. However, variable depth of soil water extraction may not necessarily be production system specific, but may be due to an insufficient period of soil drying to allow full extraction of all the available water. Hence, there is a need to use standardised procedures to quantifY the soil water supply to sugarcane in different production systems. Ratcliff et 01. (1983) found that across a wide range of soil textures, various annual crops extracted a narrow range (100-120 mm) of water per metre of soil, except for sandy soils where the value was around 60mm/m. Bridge et 01. (1982) reported the available water in Cununuura clay to be 140 mm/m. These results, and those detailed by Baran et 01. (1974), Koehler et 01. (1982) and Inman-Bamber et 01. (1999) suggest that sugarcane is capable of extracting similar amounts of water/unit soil depth as other crops. So, there is a need to quantifY both the PESW and the extent of soil water extraction with depth for specific soil types in a given location. Inman-Bamber et 01. (1998) report on the importance ofthe amount of water available to the sugarcane plant below the readily available limit in the identification of management strategies to maximise profitability rather than productivity. There are two common soil types of the Ivanhoe Plain in the ORIA, the Cununurra clay and the Aquitaine. Aldrick et 01. (1990) describes these cracking clay soils and the boundaries of the soil types. The Cununurra clay soils have a clay content of around 60% in the north and are dark greyish brown changing to reddish medium clay at 140 to 160 cm. In the south, the clay content is lower at around 45% and the soils are dark brownish changing to reddish light 49 to medium clay at 140 to 160 cm. The Aquitaine soils located in the north-east of the Ivanhoe Plain have higher clay content of around 75% and are dark grey changing to brownish medium to heavy clay at 130 to ISO cm. Development of management strategies for best practice irrigation of sugarcane requires a detailed knowledge of sugarcane water requirements. High evaporation rates are recorded in the ORIA and correspondingly high water use by sugarcane is expected. The comparison of the measured water use by sugarcane with class A pan evaporation at specific sites will enable irrigation scheduling methods to be developed to suit crop requirements. The primary objective of this section is to quantify PESW and the pattern of soil water extraction by sugarcane when grown on a range of soil types in the ORIA. A second objective is to quantify the rate of crop water use compared to class A pan evaporation throughout the crop cycle for sugarcane growing under different irrigation regimes in the Ord environment. Materials andMethods Soil water characteristics (PESW, drained upper limit (DUL) and lower limit (LL» of Cununurra clay and Aquitaine were studied on the Ivanhoe Plain of the ORIA (15.65°S, 128.nOE). Four sites were selected for study, three on Cununurra clay and one on Aquitaine soil types. The location and dates of determination are shown in Table 8.1 and the location mapped in Figure 8.1. The Central and Southern Ivanhoe Plain sites were located on brownish cracking clay. The Northern Ivanhoe Plain site is located on greyish cracking clay and the Aquitaine site has cracking clay with hydromorphic attributes and is in the greyish phase. All sites were within blocks growing sugarcane. Table 8.1. Location of the four sites studied on the Ivanhoe Plain of the Ord River Irrigation Area including the dates of determination of the drained upper limit and the crop lower limit Site Ord River Mill Field No. Southern Ivanhoe Plain BI50-1-10 Central Ivanhoe Plain 2BFWI Northern Ivanhoe Plain B220-5-20 Aquitaine B250-2-20 Drained upper limit site installed 2/5/00 115/00 415/00 5/5/00 Drained upper limit site sampled 1416/00 13/6/00 1616/00 1516/00 Crop lower limit site sampled 15/6/00, 1317100, 15/8/00, 1919/00 12/6/00, 1317100, 118/00 3118/00 2115/00, 15/8/00, 3117100 1317100, 118/00, 20/9/00 Drained upper limit After a record wet season (1662mm rainfall with the last rain on 24/4/00), 1.6 m soil moisture monitoring access tubes were installed in an uncropped area 5 m x 5 m. The site was bunded and ponded 3 times per week. Soil water content (SWC) was monitored weekly and the soil was considered to be saturated after one month, when SWC was constant. The site was covered with a black plastic sheet to prevent surface evaporation and was left to drain until readings stabilised after approximately 10 days and the profile was assumed to have drained. The soil was then sampled to 2 m using the field measurement mechanical sampling procedure for bulk density and DUL of Dalgliesh and Foale (1998). Gravimetric soil water and bulk density were determined for each 100mm depth increment using four cores at each site. 50 Figure 8.1. Ivanhoe Plain ofthe Ord River Irrigation Area Western Australia showing location ofthe soil water characterisation and crop water use sites. I \ \ 1 \ \ \, Legend Southern Inmhoe 2 Ccntrall\':lM~ 3 Nonhem l''anhoe 4 Aquitainc Highway Trock; ---; Fence -=-. Drain/channel ; Crop ""':Iter u.~~ site o_====2====i4 km 51 As Cununurra clay and Aquitaine are "shrink-swell soils", bulk density changes with soil water content. The bulk density and then the volumetric water content of samples with lower water content was calculated using the equation: p8= Pr/[Pr/Ps+8w.Pr+BJI13 (Bridge and Muchow 198 I) where p8 is the bulk density at a water content 8, pr is the fully wet reference bulk density (from DUL sites), Ps is the density of soil solids (2.64 in Cununurra clay), 8w is the gravimetric water content, B is the air filled pore space (calculated from DUL bulk density and water content). Lower limit ofwater extraction Sites for determining the lower limit of water extraction were established in early May 2000 within commercial sugarcane crops ratooned after harvest in 1999, as detailed in Table 8.1 within the same block as the DUL sites. The crops were not irrigated until after harvest except for the Aquitaine site, which was irrigated until the end of May. Evaporation from the surface was minimised by applying a thick layer of mulch. Soil water content was monitored by soil sampling at 10 em increments to 2 m depth. Three cores per site were taken at approximately monthly intervals (see Table 8.1). An indicator of crop stress was required to identify that a sufficient period of soil drying had occurred to allow the full extraction of all the available water. Crop stress was monitored by averaging the number of fully expanded green leaves on 10 stalks each week, except at Northern Ivanhoe. The crop lower limit was calculated from the gravimetric water content of the soil samples from the driest profiles and using the equation of Bridge and Muchow (1981). Crop water use Crop water use was measured at two additional sites. The first was in the sugarcane irrigation scheduling trial during the 1999/2000 growing season on the Central Ivanhoe Plain site (Table 8.1). Crops were grown under three irrigation regimes as outlined by Muchow, et al. (2001). Irrigation was scheduled after a soil water deficit of 79 (Treatment 1), 123 (Treatment 2) and 191 (Treatment 3) mm below the DUL in a 1.9 m profile. The second was in a commercial crop of plant cane Q96 (at Ord Sugar Mill Block No. 300-27-40) during the 1999/2000 growing season (Figure 8.1, site 5). Irrigation was scheduled on this cane on a regular cycle every two weeks in the absence of rain. Neutron Moisture Meter (NMM) readings were taken the day before and 2 days after each irrigation at the Central Ivanhoe site from November 1999 to September 2000. This crop was ratooned after harvest in July 1999. Four access tubes were located in each of the three replicates, and data are presented as the mean of the readings from the 12 access tubes. Gravimetric soil water samples were taken for the 0-150 mm depth and NMM readings were taken with the neutron source at 100mm intervals between 200 and 1800 mm. NMM readings were also taken the day before and two days after each irrigation from August to November 1999 within the crop planted in May 1999 at Ord Sugar Mill Block 300-27-40. Here 24 access tubes were located across the block and NMM readings were taken at the same depth increments as those done on Block 2B. Gravimetric soil water samples were taken for the 0-15 em depth. The crop water use per day was calculated as the difference between the soil water in the 1.9 m soil profile two days after irrigation and the soil water the day before the next irrigation 52 averaged over the number of days. Crop water use for irrigation intervals in which rainfall occurred was not calculated. Class A pan evaporation and rainfall were recorded daily at the Frank Wise Institute weather station. The NMM was calibrated at each site using the method of Greacen (1981). Regular volumetric soil moisture samples were taken around temporary access tubes throughout drydown cycles at each site. The soil sample and NMM readings were used to establish a calibration curve for each depth at each site. Results Drained upper limit There was considerable variation in the DUL across the four soil types at different depths in the profile (Figures 8.2b, 8.3b, 8.4b, and 8.5b). Aquitaine clay, with the highest clay content, retained the most water after drainage (1059 mm for a 2 m profile). The Cununurra clay sites retained similar total soil water in a 2 m profile (883 mm Northern Ivanhoe, 800 mm Southern Ivanhoe and 771 mm Central Ivanhoe). Soil type changes occurred at around ISO to 160 cm depth in Central and Southern Ivanhoe soils and deeper (180 cm) in Northern Ivanhoe and Aquitaine soils resulting in a change in the volumetric soil water occurring at this depth. Lower limit ofwater extraction The onset of stress was measured by the decline in the number offully expanded green leaves per stalk as the soils dried (Figures 8.2a, 8.4a, and 8.5a). The rate of water use by the crops fell during the dry-down and the sugarcane at each site extracted low amounts of water between the last two sampling dates. Accordingly, the lower limit of extraction for the sugarcane crops on these soil types was measured on a dessicated crop with little water extraction occurring. a) 850 8 800 E 0> ...: 750 .!: E .§. 700 iu j 650 '0 U) 600 ___ Soil water (mm) - r - Green lea\es 550 1May 1Jul 1Sep Date 7 z 6 "3 5~ Q. 4 . BO "0 E E 60 i 0:: 40 20 0 1-i'bv l.Jan 1-Mar 1-May 1.Ju1 1-Sep Date Figure 8_6_ Central Ivanhoe Plain (a) profile soil water to 190 cm for the ratoon crop from neutron moisture meter readings taken the day before and two days after each irrigation (1 = irrigated at 79 mm soil water deficit, 2 = irrigated at 123 mm soil water deficit, 3 = irrigated at 191 mm soil water deficit) and (b) rainfall for Frank Wise Institute 1999/2000. Drying the soil between irrigations caused the sugarcane to extract soil water to a greater depth (Figure 8.7). The NMM data at each depth for each treatment showed the extent of extraction for each treatment was different. The sugarcane extracted soil water to 60 cm depth in treatment 1, to 80 cm in treatment 2 and to 120 to 160 cm in treatment 3. 56 a) \bIUlretric Soil Moisture% 10 ZO 30 40 50 o r------"~-.-----,.-.---...----'------. 20 40 60 5 80 a'5 100 120 140 160 -.--21-Aug --e-- 25-Aug 180 -.-DUL zoo b) 10 Volurretrio Soil Moisture% 20 30 40 50 20 40 60 E 80 -"5100 0. ~120 140 160 180 200 -+-21-Aug -+-Z5-Aug ___ DUL c) 10 0 Volumetric SoilMoisture% 20 30 40 50 20 40 6D e 80 :"S 100 8' 120 140 160 -.l-21-Aug 180 -b-25-Aug -.--DUL 200 Figure 8.7. Sample profile soil water for Central Ivanhoe Plain ratoon crop from NMM readings at each depth for each treatment the day before and 2 days after irrigation and profile soil water from DUL site. a) irrigated at 79 mm soil water deficit, b) irrigated at 123 mm soil water deficit and c) irrigated at 191 mm soil water deficit. However, the rate of water extraction between irrigations was similar for sugarcane grown under the different irrigation schedules (Table 8.3). The crop water use was high before the wet season but well below the daily class A pan evaporation after the wet season. The ratio of 57 water use to pan evaporation was similar in treatment I and 2 and lower in treatment 3. This treatment was extreme and yield was reduced when irrigation was scheduled at 191mm, treatment 3 (Muchow et al., 2001), but a reduction in yield did not occur between treatments 1 and 2. Table 8.3. Central Ivanhoe Plain soil water use at each irrigation based on neutron moisture meter readings the day before and 2 days after each irrigation (12 access tubes per treatment averaged at each date) Irrigation Date Water use (mm/day) since 2 days after last irrigation Average pan evaporation (mm/day) since 2 days after last irrigation Ratio water use / pan evaporation Treatment 1 30/11199 20/5/00 30/5/00 8/6/00 117100 1517100 2717100 8/8/00 22/8/00 19/9/00 Treatment 2 10/6/00 117100 2717100 22/8/00 19/9/00 Treatment 3 22/8/00 19/9/00 6.95 1.35 2.78 2.17 1.37 2.39 2.14 2.79 2.56 3.09 2.77 2.44 2.10 2.50 2.71 1.91 2.33 6.12 1.136 5.94 0.227 5.40 0.515 5.50 0.395 5.03 0.272 4.51 0.530 5.29 0.405 5.57 0.501 5.75 0.445 7.01 0.441 5.50 0.504 5.03 0.485 5.29 0.397 5.75 0.435 7.01 0.387 5.75 0.332 7.01 0.332 The DUL of 738 mm for a 1.9 m profile obtained after extended ponding at the DUL sample site was not reached in the sugarcane cropped area, even after the extended wet season (Figure 8.6). The extent of refill after irrigation was low and differed for each treatment with the drier treatments being drier at depth (Figure 8.7). The DUL was reached in the surface 60cm, but irrigation water applied had little effect on the lower depths especially in treatment 3. The average soil water after irrigation in treatment I was 73% of the PESW compared to 64% in treatment 2 and 59% in treatment 3. Chan (1981) found that the maximum storage capacity of the soil is rarely achieved under furrow irrigation of cracking clays. Deep drainage losses to the groundwater are minimal under normal irrigation practices in Cununurra clay at this site. The profile soil water at Block 300-27-40 (Figure 8.8) shows that the amount of soil water stored at irrigation in a commercial crop scheduled for irrigation every 2 weeks varied through the season. From NMM readings at each depth, this vigorous plant crop extracted moisture to 160 cm and was drying out at depth as irrigation was not replacing all the soil water extracted below 60 cm (Figure 8.9). Rainfall prevented further measurements. The crop water use was low in early August in the young crop in cooler temperatures, but the crop 58 began to use water at rates similar to pan evaporation in late August, September and October (Table 8.4). 780 E '" .£ 730 Q:; 1Il l: '0 E'" 680 E ro ~ 630 N 580 1.Jul 1-Aug 1-Sep 2-Oct Date 2-Nov 3-0ec Figure 8.8. Profile soil water to 170 cm for Ord Sugar Mill Block 300-27-40 plant crop from neutron moisture readings taken the day before and two days after each irrigation 1999. a) Volumetric SoIl Moisture % 10 20 30 40 50 O,---~--~-~-~ 20 40 60 E60 :"5100 8"-120 140 160 ___ 14JuI 180 --e-17-JuI 200 b) Volumetric Soil Moisture % 10 20 30 40 50 O,--_-~--~-~ 20 40 60 E 80 "J::: 100 ! 120 140 160 -ll--11-0ct 180 ---a- 14-0ct 200 Figure 8.9. Sample profile soil water for Ord Sugar Mill Block 300-27-40 plant crop from neutron moisture readings taken the day before and two days after irrigation (a) 14 July 1999 and (b) 11 October 1999. 59 Table 8.4. Ord Sugar Mill Block 300-27-40 soil water use at each irrigation based on neutron moisture meter readings the day before and 2 days after each irrigation (24 access tubes averaged at each date) Irrigation Date 16/8/99 30/8/99 13/9/99 27/9/99 1111 0/99 Water use (mm/day) since 2 days after last irrigation 3.3 6.3 7.5 6.8 7.4 Average pan evaporation (mm/day) since 2 days after last irrigation 6.10 6.67 6.84 7.24 7.64 Ratio of water use / pan evaporation 0.541 0.945 1.097 0.939 0.969 Large differences between crop water use measured before and after the wet season occurred at the two sites. The ratio of water use to pan evaporation changed through the crop growth season, with high water use in a young growing crop in the pre-"wet" season and lower crop water use in a mature crop in the post-"wet" season before harvest. Discussion This study has provided information on the plant extractable soil water and pattern of water extraction with depth in the soil profile for a range of soil types in the ORIA. Four broad soil types across the Ivanhoe Plain were characterised. At the Central Ivanhoe site, 226 mm of water in 2m profile was available to the sugarcane plant. The plant extractable soil water in 2 m differed at the other sites with the Aquitaine, Northern Ivanhoe and the Southern Ivanhoe soils holding 187, 193 and 167 mm respectively. The depth of water extraction varied across sites with extraction occurring to 1.8 m at the Central and Northern Ivanhoe sites, 1.6 m at the Southern Ivanhoe site and to only I m at the Aquitaine site. The four soil types are different in PESW and depth of extraction and so recommendations for scheduling irrigation across the ORIA using the soil water deficit as an indicator of when to irrigate is not accurate. Differences in PESW among soil types could lead to differences in the optimum soil water deficits at which to irrigate. The optimum soil water deficit at which to irrigate developed at one site can be converted to fraction of extractable soil water (FESW) remaining in the soil before irrigation. This figure can be transferable to other sites if the DUL and LL of these sites are known. For example the soil water deficit at the Central Ivanhoe site of 120mm is approximately equivalent to 0.5 FESW. The FESW of 0.5 at the Southern Ivanhoe site is equivalent to 84mm soil water deficit. Knowledge of the PESW across the ORIA will aid researchers to develop recommendations for irrigation scheduling across the ORIA. The standardised procedures used in this report will enable successful comparisons to be applied across further soils in the ORIA. High rates of crop water use occurred in a sugarcane crop before the wet season, but low rates of water use relative to class A pan evaporation were observed after the wet season. This has implications for irrigation scheduling in the ORIA. Matching irrigations to crop water requirements will improve crop water use efficiency and reduce losses to the environment. Knowledge of the extractable soil water and crop water requirements allows a better understanding of the response of sugarcane to irrigation schedules for different soil types. 60 The soil water characteristics information for these soil types in the ORIA is a prerequisite for the use of cropping system models to analyse the conseqnences of different irrigation regimes to develop strategies for best-practice irrigation management that maximise profitability and irrigation efficiency. EnviroSCAN investigations EnviroSCAN equipment was extensively used in an earlier SRDC funded project in the Ord for assessing drainage below the root zone (Nulsen and Sherrard, 1999). Firstly the equipment was calibrated for a range of soil types, as described in Sherrard et aI., (2002). An EnviroSCAN system was then installed in a first ratoon crop of sugar cane on Block 55 in 1996. The soil moisture level in this crop was monitored continuously until harvest in September 1997. After harvest, cultivation and fertiliser application, the system was re-installed in this same crop and changes in soil moisture further monitored until harvest in August 1998, to observe the possible OCCUrrence of deep drainage. The same investigation also indicated reduced soil moisture extraction by the sugarcane crop as the wet season progressed. In this project, EnviroSCAN equipment has largely been used for assessing crop water use. In Block 27 soil moisture levels were monitored for the 1999-2000 and 2000-2001 crops to record daily changes in soil moisture. These data were used to determine crop water use as well as to assess any drainage below the root zone. This allowed determination of water extraction on a daily basis and for the entire crop season (Figures 8.10 and 8.11). Due to confounding influences of wet season rainfall with saturated profiles, soil moisture extraction was assessed between periods of rainfall and extrapolated to estimate extraction on a daily basis. 12 -~ 10 "0 E E8 -"'":::J ~ 6 "~ 4 co- ~ tl 2 -1999-2000 (P) 0 May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Figure 8.10. Soil moisture extraction by a plant crop of Q96 grown through the 1999-2000 season. The crop was planted in May 1999 and was harvested in June 2000. 61 12 i;' 10 ~ .E 8 ""~ 6 S ~4 eQ. () 2 -2000-2001 (1R) ~-~"~~"---~ "---~- -~--~-~ o~~~~~~--~--~~--~--~~--~~--~--~~ May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Figure 8.11. Soil moisture extraction by a first ratoon crop of Q96 grown through the 20002001 season. The crop was ratooned in June 2000 and was harvested in June 2001. Water extraction peaked at 10.9mm/day for the 1999-2000 crop but was only slightly lower for the 2000-01 season. By integrating daily water use, total crop water use for the 1999-2000 and 2000-2001 crop cycles was estimated to be 17.3 and 13.5 ML/ha respectively. With crop yields of 183 and 140 tonnes cane/ha, 10.8 and lOAt cane were produced per ML water used for the 1999-2000 and 2000-2001 crops respectively. High rates of water use occurred for both crops prior to the wet season, dropping significantly as the wet season progressed and well before harvest. This has important implications for irrigation scheduling in the ORIA and for water allocation planning. Matching irrigation frequency and duration to crop water requirements will improve crop water use efficiency and reduce losses to the environment. This work also supports the finding of a growth slow-down observed in the sugarcane irrigation trial on Block 2B. 62 9. WEATHER CONDITIONS IN THE ORD FOR THE DURATION OF THE PROJECT One of the main reasons for conducting this irrigation project in the Ord is that weather conditions are quite different from those in Queensland where most of the previous irrigation research has been conducted. It was important to compile credible current and long-term climate records in order to interpret the results of the field experiments and to extrapolate the results to different weather conditions that may occur in future years. Two Campbell automatic weather stations were purchased using SRDC funds and were installed and calibrated. One was located at Cummings Brothers' farm, near the Frank Wise Research Station and the other at George Gardiner's farm at Packsaddle, in the southern part of the Ord River Irrigation area. The weather stations were installed in November 1996 and were working satisfactorily in March 1997. In addition to the two automatic weather stations, the manual weather station at the Frank Wise Research Institute was thoroughly checked, the radiation sensor recalibrated and the weather data obtained after 1988 checked for errors. Rainfall Annual rainfall in the Ord is extremely variable with less than 500 mm falling in some years and over 1400 mm falling in other years. Most of the rain falls from December to May. Rainfall in June to August is negligible (Figure 9.1). 1500 E E ~ .cJ:.!!1000 - .~ I I • Cii :J § 500 ;- I ring II - t·-1' I I-I j I 1·.1·· I. I .1-- II II ..~ I· I ,yA tumn - .....-s mmer 1965 1975 1985 1995 2005 Figure 9.1. Contribution of rain in each season to total annual rainfall in the Ord. Annual rainfall increased during the project from 621 mm in 1998, to 1286 mm in 1999, 1366 mm in 2000 and 1485 mm in 2001. Thus the project was conducted during some of the wettest years of the past 45 years. No other period in the past 45 years has experienced three such wet years in a row as 1999 to 2001. The distribution of rainfall during the project is shown in Figure 9.2. A characteristic of the wet season in each year was its abrupt end whereas the start of the wet season was not clearly identified in the rainfall record. This means that irrigation scheduling during the start of the wet season will be complicated by rain but after the wet season the chances of disruption to any schedule will be low. Daily rainfall exceeded 100 mm on occasions and often exceeded 50 mm. Most of these high rainfalls would have been ineffective when soils were saturated from preceding frequent rain. 63 15n--------.--------,--------.--------, 1998 1~~--------.--------,--------.--------, 1 1999 ~ E E ~ ~c 1 .~ Z, Oro 1 0 2000 2001 o 100 200 300 400 Day of year Figure 9.2. Daily rainfall measured at the Kimberley Research Station for 1998-2001 64 Radiatioll Total annual radiation at the Kimberley Research Station is shown in Figure 9.3. The radiation during the life of the project was not particularly high compared to previous years as may be expected from the high rainfall during 1999 to 200 1. Radiation was generally lowest in summer and highest in autumn. This may also be a distinguishing feature of the Ord climate in terms of crop growth. Low radiation combined with high temperature may have had unexpected effects on the crop such as significant losses of biomass due to respiration. 9000 ~ '"E 8000 r-- --., -::2: 7000 ~ :II IIc 6000 11111I111111111I111111 - SpriJ 9 11111I111I111111 - 0 +rn' 5000 1-' '0 ~ 4000 I- Aub: mn - r~n 3000 I- 0 C 1000 I- 0 - Wi!! er - SUIT mer 1955 1965 1975 1985 1995 2005 Figure 9.3. Contribution of solar radiation in each season to total annual radiation in the Ord. Daily radiation for the years of the project are shown in Fignre 9.4. The variation in radiation throughout the year is low compared to more southerly locations where cane is grown and it is noteworthy that winter or dry season radiation can be considerably greater than radiation in the wet season (Figure 9.3) even though peak radiation is higher in the wet season. There are two periods of peak radiation corresponding with two periods of maximum solar elevation between the equinoxes and summer solstice (Figure 9.4). 65 4~-------'---------'-------'r-------, 1998 3 1 oo o o o 4~====~====~====~====~ 2000 3 ~~oo~~~_~~&o~'2 00 ~ ~~"'{!PVO- 0 ~ 00> '0 OJ 0 @o 0 00 (§l 0 1o 0 o0 08 o 4 o0 3 ~,.g;iEo 0 0 2 o ~o £:,'8 «)'bo c9 S 1 o0 0 0 0 0 0 2001 00 oo o o 100 200 300 400 Day of year Figure 9.4. Daily solar radiation measured at the Kimberley Research Station for 1998 to 2001 66 Temperature Mean winter or dry season temperature varied considerably from year to year exceeding 25°C in some years and dropping below 22°C in other years (Figure 9.5). Mean summer (wet season) temperature has been remarkably consistent over the past 45 years ranging from 30 to 32 °C in all but 5 of the 45 years. The variation in daily maximum temperature during the years of the project was not great (Figure 9.6). Maximum temperature varied between 30 and 40°C for more than 80 % of days from 1998 to 2001. In about 5 % of days maximum temperature exceeded 40 0c. Not much is known about growth of sugarcane under these conditions. Extended periods of high temperature (> 35°C) and occasional periods of extremely high temperature (>40°C) is what distinguishes the Ord region from other sugarcane regions in Australia. Minimum temperature varied more than maximum temperature over each of the years of the project (Figure 9.6). There were several days when minimum temperature was less than 10°C and it was less than 20 °C for more than 35 % of days. 35,-______- .________. -______- .______- .______--. ~ :::J Ti1 OJ 30 Q. -E OJ Auumn crn: OJ E 25 rn oercnn: 20L-______OJ (J) - L_ _ _ _ _ _ _ _L -_ _ _ _ _ _- L_ _ _ _ _ _ _~~ _ _ _ _ _~ 1955 1965 1975 1985 1995 2005 Figure 9.5. Mean temperature for each season over a 45 year period at Kimberley Research Station 67 5 4 3 2 1 5 U 0 4 ~ -~ :J 3 .C..\.l (!) 2 c.. - 1E (!) E :J E5 :~ E4 "0 C C\l 3 E :J ·Ex 2 C\l 1 E ·zrn- 0 1998 1999 2000 2001 o 100 200 300 400 Figure 9.6. Daily maximum and minimum temperature measured at the Kimberley Research Station for 1998 to 2001 68 10. DEVELOPING IRRIGATION GUIDELINES USING THE APSIM SUGARCANE MODEL Illtroductioll Sustainable irrigation management needs to take account of the productivity and profitability responses to irrigation timing and amount, and concurrently the need to maintain water resources and minimise losses and off-site environmental consequences. In developing sustainable irrigation practice, the first step is to quantify the yield response to different irrigation schedules. Previously, research on sugarcane has been conducted in the Ord during the periods 1951-57, 1964-68 and 1975-82 (Cox and Chapman 1985). With ample low-priced irrigation water, Iittle attention was paid to irrigation practice, but the few irrigation trials that were conducted gave no significant difference among irrigation treatments (Kingston et al. 1980). In the absence of data on Ord specific irrigation responses, Muchow and Keating (1998) used a modelling approach based on knowledge from other parts of the world to develop indicative estimates of irrigation water requirements and yield consequences for different management options for sugarcane production in the Ord. This analysis indicated an irrigation requirement up to 23 MLlha, but the authors cautioned that field experimentation was a requirement under Ord conditions to confirm these estimates. The necessary field experimentation was carried out as reported in section 7 of this report. From these experiments alone, it is difficult to develop irrigation scheduling guidelines for the Ord sugar industry. However, these experiments do allow the testing of the modelling approach and indicative estimates of irrigation experiments as provided by Muchow and Keating (1998). To proceed it was necessary to: (i) understand the processes of yield accumulation in these experimental crops in response to water supply; (ii) encapsulate this understanding into a sugarcane simulation model; and (iii) use the simulation model to develop irrigation guidelines. Accordingly, the aim of this section is firstly to test the APSIM Sugarcane model (Keating et al. 1999) against observations of yield accumulation and soil water extraction from Field 2A experiments reported in section 7. Secondly we allowed for a process of calibration, if this was required, to tailor the model for Ord conditions. The third aim was to link long-term climatic data and soil water characteristics of the main soil type (Cununurra Clay) with APSIM Sugarcane to develop indicative estimates of irrigation water requirement and yield consequences for different irrigation scheduling options for sugarcane production in the Ord. Materials andMethods Field and model analysis ojyield accumulation and soil water extraction Yield accumulation in two crop cycles was measured for crops grown on Cununurra Clay (Bridge and Muchow 1982) at the Frank Wise Institute Kununurra (lat. 15.65° S, long. 128.72° E). A plant (P) crop was planted on 6 May 1998 and the 1st ratoon (R) crop was started on 4 August 1999 (see Section 7). Three irrigation treatments with 3 replications were applied to each crop as outlined in Section 7 and also by Muchow et al. (200 I). The P crop was irrigated after 120 mm of pan evaporation until canopy closure on 18/9/98, and thereafter irrigated after either 60,120 or 180 mm of pan evaporation. The ratoon crop was irrigated after 120 mm of pan evaporation until canopy closure on 15/11/99, and thereafter irrigated after either a soil water deficit of 79, 123 or 191 mm. The crops were furrow irrigated. For 69 the P crop, a drying-off period prior to harvest of 26 days was imposed for the 60mm pan evaporation treatment, 27 days for the 120mm pan evaporation treatment and 28 days for the 180mm pan evaporation treatment. For the R crop, a drying-off period of 28 days was imposed on all treatments prior to harvest. The crops were hand-sampled with harvesting 18 m' of crop at intervals as shown in Figure 10.1. Crop biomass and stalk sucrose were determined using the procedures outlined in Muchow et al. (1993). These data are from clean stalk samples and differ from the commercial yields obtained by mechanically harvesting large plots as shown in Muchow et al. (2001). Soil water extraction was measured throughout crop growth using a Neutron Moisture Meter, as outlined by Muchow et al. (2001). Simulation analysis ojirrigation strategies The APSIM-Sugarcane model is designed to simulate a uniform block of cane and predict on an area basis, crop biomass, sucrose yield, crop water use and irrigation requirement, assuming 100% irrigation application efficiency (Keating et al. 1999). The model operates on a daily time-step and is 'driven' by climatic factors including rainfall, temperature and solar radiation; soil factors such as depth, water holding capacity and nitrogen status; management factors such as planting/ratooning date, fertiliser and irrigation; and genetic factors. The soil water module for APSIM is described in Probert et al. (1998). The model has not been tested previously in the Ord environment because of the lack of complete datasets where crop, soil, climate and management data were collected. The analysis here provides the first test of APSIM-Sugarcane in the Ord environment. All simulations were conducted for Cununurra Clay having soil water characteristics as shown in Table 10.1. Muchow et al. (2001) observed that sngarcane extracted water to 1.9 m on this soil type, and the total plant extractable soil water (PESW) was 225 mm. Plant extractable soil water is defined as the water held at tensions between the crop lower limit and the drained upper limit, parameters that are determined empirically using methods described by DalgJiesh and Foale (1998). Nutrition was assumed to be non-limiting. Stage 1 soil evaporation coefficient was set to 6 mm, stage 2 soil evaporation was set to 3.5 mm and curve number was set to 82 (Muchow and Keating 1998). The effectiveness of the canopy on reducing curve number and the maximum reduction in curve number due to cover were set to zero. The crop transpiration efficiency was set to 0.01 and the coefficient for the vapour pressure deficit was changed from 0.75 to 0.70 to represent Ord conditions based on the analysis of B.A. Keating and N.!. Ruth (unpublished data). Table 10.1. Characteristics of Cununurra Clay soil from Muchow et al. (2001) used in APSIM-Sugarcane SlmuIatl.On runs Depth (mm) Bulk Density Drained upper limit Lower limit 0-150 150-250 250-350 350-450 450-550 550-650 650-750 750-850 850-950 950-1050 1050-1150 1150-1250 1.502 1.497 1.513 1.509 1.466 1.510 1.519 1.513 1.548 1.495 1.435 1.462 0.392 0.368 0.375 0.380 0.379 0.392 0.402 0.393 0.403 0.393 0.379 0.397 0.170 0.234 0.269 0.274 0.267 0.277 0.283 0.290 0.303 0.301 0.294 0.303 70 1250-1350 1350-1450 1450-1550 1550-1650 1650-1750 1750-1850 1850-1900 1.512 1.539 1.562 1.627 1.708 1.708 1.721 0.415 0.421 0.412 0.398 0.372 0.345 0.338 0.304 0.270 0.251 0.266 0.267 0.277 0.300 The model analysis approach requires quality climatic data, soil water characteristics and confidence in the ability ofthe model to simulate responses to specified inputs. For the model analysis of the two experiments of Muchow et at. (2001), daily radiation, maximum and minimum temperature and rainfall data were sourced from an automatic weather station sited 0.5 km from the experiments. Using these climate data and the observed soil water characteristics ofthe experimental site (Table 10.1), simulations were conducted using the observed crop start and finish dates and the observed dates of furrow irrigation. The extent of refill of the soil profile at each irrigation in the simulations was based on observed soil water profiles after irrigation. For the P crop, the profile was filled to 80% of tota! plant extractable soil water (PESW) at irrigation. For the R crop, the profile was filled to 78% of total PESW for the 79mm treatment, 73% for the 123mm treatment and 70% for the 191mm treatment. Those differences reflect limitations in rates of water entry and re-distribution in the heavy clay soils of the region. In developing irrigation scheduling guidelines, long-term climatic data (daily solar radiation, maximum and minimum temperatures, rainfall and Class A pan evaporation) were sourced from the Frank Wise Institute for the period 1957 to 2000. For the long-term simulations, APSIM Sugarcane was configured for a ratoon crop starting either on I June, I August or I October and harvested at 12 months of age. The initial soil water was set to a FESW of 0.1. Irrigation was scheduled the day after ratooning and second irrigation was scheduled at 30 days after ratooning in the absence of rainfall. Thereafter, irrigation was scheduled at different soil water deficit as shown in Table 10.2 with a drying-off period prior to harvest of 30 days. Irrigation was assumed to fill the profile to 75% of total PESW and the amount of irrigation applied assumes an application efficiency of 100%. Table 10.2. Number of irrigations, irrigation amount (at 100% application efficiency) and sucrose yield for irrigation at different soil water deficits with a 30 day drying off period, averaged for years 1957-2000 for ratoon crops started on I June, I August and I October and harvested at 12 months. The PESW was 225 mm and irrigation was scheduled at Fraction Extractable Soil Water (FESW) ranging from 0.6 to 0.1. Irrigation Strategy FESW Deficit (mm) Variable 0.6 90 N\!.· irrigations 0.5 112.5 " 0.4 135 " 0.3 157.5 " 0.2 180 " 0.1 202.5 " 0.6 90 Irrigation (mm) 0.5 112.5 " 0.4 135 " 0.3 157.5 " 0.2 180 " 0.1 202.5 " 1 June 36.9 23.0 16.7 13.0 10.1 7.3 1499 1448 1398 1319 1192 948 Crop Start 1 August 34.2 21.1 15.4 11.9 9.3 6.9 1384 1324 1275 1201 1088 891 1 October 30.7 19.0 13.7 10.9 8.8 6.7 1222 1180 1132 1098 1025 878 71 0.6 90 Sucrose yield (Uba) 20.6 20.3 20.1 0.5 112.5 " 20.5 20.2 20.0 0.4 135 " 20.2 20.0 19.8 0.3 157.5 " 19.5 19.4 19.6 0.2 180 " 18.! 18.4 18.8 0.1 202.5 " 15.5 16.4 16.7 Results and Discussion Yield accumulation and soil water extraction The accumulation of aboveground biomass over time is shown for the P and R crop in Figure 10.1. There was a marked reduction in growth rate after sampling 2 (224 days after planting or 16 December 1998 and 160 days after ratooning or I January 2000). For comparison, a plant crop grown in the Burdekin under fully irrigated conditions in the 1991-92 season and similarly sampled (Muchow et al. 1994) is shown in Figure 10.1. This Burdekin crop showed a linear growth rate from 214 to 353 days after planting of 41.3 g m-' d-l For the P crop grown in the Ord, the growth rate between sampling 1 and 2 (153 to 224 days after planting) was 30.4 g m-' d-\ and thereafter the average growth rate was 10.1 g m-' dol. A similar trend was evident for the R crop in the Ord (Fignre 10.1). The growth rate between sampling 1 and 2 (97 to 160 days after ratooning) was 28.5 g m-' dol, and thereafter the average growth rate was 4.9 g m-' dol. 80.---.----.---,----,---, ~, 70 .rco ~60 • Ord P • Ord R .... Burdekin P CIl CIl ~ 50 o :0 40 -0 C e::J 30 OJ ID 20 ~ ~ 10 100 200 300 400 500 Days after planting/ratooning Figure 10.1. Accumulation of aboveground biomass (dry weight) over time for the plant (Ord P) and ratoon (Ord R) crop grown in 1998-99 and 1999-00 respectively in the Ord. Data for the Ord P crop are the mean of the 60mm and 120mm pan evaporation treatments of Muchow et al. (2001). Data for the Ord R crop are the mean of the 79mm and 123mm soil water deficit treatments of Muchow et al. (2001). A plant crop grown in the Burdekin (Burdekin P) under fully irrigated conditions in the 1991-92 season and similarly sampled as reported by Muchow et ai. (1994) is also shown. 72 The factors contributing to this growth reduction from mid-wet season in the Ord are not known. However, the crops did lodge during this period, and this may be a cause. A slowdown in growth has been observed for large lodged crops by Muchow et al. (1995), Robertson et al. (1996) and Muchow et al. (I 997a). APSIM-Sugarcane estimates of crop biomass in the ahsence of lodging (broken line) tended to be much higher than the observed crop biomass, particularly during later growth (Figure 10.2). This is in contrast to the many environments where APSIM Sugarcane has been shown to reliably simulate field performance (Muchow et al. 1997b; Lisson et al. 2000). The simulation model of Keating et al. (1999), as used in the indicative analyses of irrigation requirements in the Ord by Muchow and Keating (1998), was not configured for any growth reduction that may possibly be attributed to lodging. Based on the analysis of radiation use efficiency of these crops (Muchow et al. unpublished data) and the observations of Singh et al. (2000) for lodged crops, the model was re-configured to reduce radiation use efficiency (RUE) to 1.0 g/MJ after above ground biomass reached 30 lIha, since this was approximately the above-ground biomass at sampling 2 (Figure 10.1). With these modifications, APSIM Sugarcane simulated less growth during the later growth stages (solid line, Figure 10.2) and closely mirrored the observed growth patterns in the field grown P and R crops (Fig. 2). Pooling all data, the simulations using the modified APSIM-Sugarcane model were in close agreement with observed crop biomass over the range of 10 to 60 lIha (Figure IO.3a). There was a slightly lower correlation in the simulation of sucrose yield, but still more than 90% of the variation was accounted for (Figure 10.3b). The pattern of soil water extraction also closely followed the observed data, especially for the plant crop (Figure 10.4). 73 a) P Irrig at 60mm Pan evap // ~ 80 • Observed // / "-", I .r:: I "~" 60 :Eoc 40 I I I I I e"- I () 20 d) R Irrig at 79mm SWD 80 • Observed "", .r:: ~ 60 :E""o'a"'' 40 e"- () 20 / / / I I I I I _OL--LL-~L-~~ _L-~ o 100 200 300 400 500 Days after planting 100 200 300 400 500 Days after ratooning 100.---,----,----,---,----, b) P Irrig at 120mm Pan evap ~ 80 "-", .r:: e. 60 ~ :oEc 40 e- () 20 / / / I / I / I I / 100 200 300 400 500 Days after planting 100.---.----.---,----,---, e) R Irrig at 123mm SWD ~ 80 "", .r:: ":E""'oa""'' 60 40 e"- () 20 • / I / / I I / • • / 100 200 300 400 500 Days after ratooning c) P Irrig at 180mm Pan evap ~ 80 'O'" .r:: ~ 60 "l{'l E :§ 40 "e- () 20 ,.- / I ./ / I / • °O~~L---~--~--~--~ 100 200 300 400 500 Days after planting 100.---,----.---.----,---, f) R Irrig at 191mm SWD ~ 80 .'r":: .:t::- 60 .""'§"'' .0 40 e- () 20 • • / 100 200 300 400 500 Days after ratooning Figure 10.2. Observed and simulated crop biomass (dry weight) over time for the P crop for irrigation after a) 60, b) 120 and c) 180 mm of pan evaporation and for the R crop for irrigation at a soil water deficit of d) 79, e) 123 and f) 191 mm. The simulated crop growth (broken line), using the original APSIM Sugarcane model of Keating et a!. (1999), and the simulated crop growth (solid line), using the model configured with the growth slow-down phenomenon, are also shown. 74 ~ 70 ~ 'ro aJ .c 60 ~ (J) gJ 50 E ~ 40 / • .~• ;,,'\, •• • g"- 30 "0 ~lzo ::; i:Eii 10 ..... .- ",-,- " •• )",1.01 x-0.834'.0.940 / 0a 10 20 30 40 50 60 70 Observed crop biomass (t ha-1) ~30 ~ 'ro bJ .c :t::... 25 g'"(J) 20 :;, (J) ~ 15 2 -(J) 210 :r:o; .E i:ii 5 ,- A' / / .. ." / / •• ~/. ...,• ..& .-. •••• • • • y=0.984x-O.15hO.911 00 5 10 15 20 25 30 Observed slalk sucrose (I ha-1) Figure 10.3. Simulated (using the model configured with the growth slow-down phenomenon) vs observed a) crop biomass (dry weight) and b) stalk sucrose for all P crop treatments and all R crop treatments. By using the original model of Keating et al. (1999) and the modified model, the significance of a lower RUE during later growth can be assessed. The growth "slow-down" phenomenon substantially impacted on yield. For example, at the final sampling at 449 days after planting of the P crop irrigated after 60 mm of pan evaporation, the observed crop biomass was 58.2 tlha, and the simulated crop biomass configured for growth slow-down was 59.9 tlha. Yet the simulated crop biomass using the original model without growth slow-down was 96.0 tlha (Fig, 2). The corresponding observed sucrose yield was 24.2 tlha compared with the simulated sucrose yield with growth slow-down of 24. I tlha. However, the potential sucrose yield for the 15-month-old crop with no growth slow-down was 40.5 tlha. The growth slow-down phenomenon also impacts on crop water use, with the irrigation requirement (at 100% application efficiency) for this crop decreasing from 2045 to ISIS mm. The modified model was used in the subsequent simulations presented in this paper as it closely describes the observed crop ~wth, yield and water use of the range of irrigation treatments over two seasons. Further field investigation of the growth slow-down phenomenon for different seasonal conditions and crop start (ratooning) dates is required, together with studies of how best to represent this phenomenon in cropping models. In the meantime, the simulation analyses that follow provide the current best-bet estimates of sugarcane crop responses to different irrigation schedules. 75 ~--------------- -.-- - - - - - - - - - - - - - "0 (f) ..2cl 100 j!j § 50 • Observed - - Simulated oL-~_-L_~_L-~_~ 370 380 390 400 410 420 430 Days after planting b2 ______________ _ 1200 .li ~ 150 "0 (f) ..2cl 100 j!j ;~lj 50 • • _ _ _oL-~ _L_~_L-~ ~ 370 380 390 400 410 420 430 Days after planting ~--------------- E.§. 200 ~ .2l ~ 150 • "0 (f) ..2cl 100 {!l § 50 • _ _oL--L_~~ - L_ _L-~ 370 380 390 400 410 420 430 Days after planting 250 d) 1200 ~ .2l ;\: 150 ~ (f) ..2cl 100 j!j x~ 50 W • Observed - - Simulated 0 200 250 300 350 400 Days after ratooning 250 r-----,---,-----,------, e) .E§. 200 ~ .2l ;\: 150 "0 (f) !..2cl 100 50 O~---L--~-~----~ 200 250 300 350 400 Days after ratooning 250r----.---,---.....- - - , .E§. 200 .$ ;\: 150 ~ I..2cl 100 50 •• _OL---~ ___~____L - _ - " 200 250 300 350 400 Days after ratooning Figure 10.4. Soil water extraction over time for the P crop for irrigation after a) 60mm, b) 120mm and c) 180 mm of pan evaporation and for tbe R crop for irrigation after d) 79mm, e) 123mm and f) 191mm soil water deficit. The PESW of 225 mm and the refill points at irrigation (80% for a), b) and c), 78% for d), 73% for e) and 70% for f) are shown by the parallel broken lines" 76 Simulatedyield and irrigation response to soil water deficit Crops were simulated with the modified APSIM Sugarcane model starting in the 1957-58 season and finishing in the 1999-2000 season. The average sucrose yield response to irrigation at different soil water deficits for ratoon crops started on I June, 1 Augnst and 1 October is shown in Table 2. As expected the number of irrigations decrease markedly when the irrigation interval is extended from irrigating at a soil water deficit of 90 mm to 135 mm; thereafter the number of irrigations was not markedly changed for irrigation at greater soil water deficits. The amount of irrigation applied (assuming 100% irrigation application efficiency) progressively decreased with greater soil water deficit at irrigation (Table 10.2). Application efficiency would be lower with more frequent irrigation as greater soil water deficits at irrigation result in more soil cracking and better infiltration (Muchow and Wood 1981). Consequently, the actual amounts of irrigation applied in field practice would be greater than the data shown in Table 10.2, especially for the more frequent irrigation schedules. Crops starting on 1 June required most irrigation while those starting on 1 October required least irrigation. Muchow and Keating (1998) used historical climate for 1960-1985 and for a crop ratooned on 1 September and irrigated at FESW of 0.5, estimated an irrigation requirement of 2269 mm applied in 22 irrigations and a potential sucrose yield of 29.0 tlha. These results did not account for the growtll reductions being observed during the wet season in large lodged crops in the Ord. In contrast in the current analysis, August and October ratoons irrigated at FESW of 0.5 respectively had an irrigation requirement of 1324 to 1180 mm applied in 21-19 irrigations, associated with simulated sucrose yields of 20.2 to 20.0 tlha (Table 10.2). The growth - slow down phenomenon reduces the irrigation requirement by approximately 40% and reduces sucrose yield by up to 30%. The sucrose yield response to irrigation schedule shows relatively minor effects on yield until the soil water deficit at irrigation is extended beyond 135 mm (Table 10.2). In contrast to the number and amount of irrigation, sucrose yield was least sensitive to crop start date. The most profitable irrigation schedule will be dependent on the cost of water as determined by amount applied, labour costs as influenced by the number of irrigations and the price for sugar. However, data as shown in Table 10.2 are valuable for examining the tradeoffs with different irrigation schedules. Given the need for more field-testing of the model, we would advocate a conservative approach at this stage and subsequent analyses are present for irrigation scheduling at a soil water deficit of 135 mm. Year to year variability in yield and irrigation requirement The year-to-year variation in irrigation requirement when irrigation was scheduled at a soil water deficit of 135 mm for the last 10 years is shown in Table 10.3. Rainfall during the wet season is highly variable from year to year. In fact for the period 1957 to 2000, the driest (1997-98) and wettest (1999-2000) year occurred in recent times. For the driest year with 473 mm of rainfall, 19 irrigations were required to apply 1573 mm of irrigation whereas in the wettest year with 1443 mm of rainfall, 9 irrigations were required to apply 760 mm of irrigation. More of the rainfall was effective in the driest year. Sucrose yield was higher in 1997-98 than in 1999-2000, most likely associated with higher solar radiation from less cloud cover. However, over the historical record, there is little association between sucrose yield and amount of rainfall. 77 Table 10.3. Yearly variation in rainfall, effective rainfall (%), nnmber and amount of irrigation and sucrose yield for irrigation at 0.4 FESW (135 mm soil water deficit) with a 30 day drying-off period for 12 month ratoon crops started on 1 Angust for years 1990-1991 to 1999 -2000 Season 1990-91 1991-92 1992-93 1993-94 1994-95 1995-96 1996-97 1997-98 1998-99 1999-2000 Rainfall Jrnm) 596 578 783 722 1143 826 1061 473 1145 1443 Effective rainfall (%) 70.2 77.1 59.1 58.2 61.3 73.6 56.1 82.6 66.6 51.0 Number of Irrigations 17 18 17 16 12 15 15 19 12 9 Amount of Irrigation (rnm) 1407 1490 1401 1318 996 1236 1239 1573 1003 760 Sucrose yield (!/hal 19.5 20.3 19.9 19.7 20.3 20.5 20.1 20.4 19.2 17.1 Table 10.3 highlights the importance of distribntion of rainfall given that the nnmber of irrigations is not closely related with rainfall. For example, more irrigations were reqnired in 1992-93 with a rainfall of 783 mm than in 1993-94 when rainfall was less at 722 mm. This resulted in a difference in irrigation requirement from 1401 to 1318 mm. The irrigation requirement also varied with crop start date. For the driest year (1997-98) a 1 June start date reqnired 20 irrigations' applying 1655 mm of irrigation, whereas a 1 October start date reqnired 18 irrigations applying 1500 mm of irrigation (data not shown). Hence, the greatest source of variation in irrigation reqnirement is year-to-year variation rather than different cropping cycles. Comparison of the irrigation reqnirement for 1995/96 (Table 10.3) with the benchmarking survey of actual practice (Wood et al. 1998) shows that considerable savings can be made in irrigation in the ORIA. The simulations in Table 10.3 indicate a requirement for 15 irrigations and 1236 mm of irrigation in 1995-96. In the survey, the average number of irrigations was 27 and the annual irrigation applied at 70% application efficiency was 3250 mm, which translates into a requirement of 2175 mm at 100% application efficiency. Whilst there is considerable variation in the survey benchmark data of Wood et al. (1998) and not all crops were initiated in August and were harvested at 12 months of age, nevertheless the difference in irrigation requirement is large. This indicates considerable potential for more efficient use of water reSOurces for sugarcane production in the ORIA. Monthly irrigation scheduling guidelines The scheduling of irrigation throughout the crop cycle will depend on the extent of canopy development, the rate of crop growth, the potential evaporative demand and rainfall patterns. To guide irrigation practice on-farm, an analysis was undertaken on a monthly basis to determine the number of irrigations required per month and the trigger point for each irrigation based on Class A pan evaporation (Table lOA). The analysis was undertaken using a conservative irrigation strategy of irrigating after a soil water deficit of 135 mm for the driest season (1997-98) and the wettest season (1999-2000) where the total number of irrigations reqnired ranged from 19 to 9 (Table 10.3). Given an irrigation schedule that triggers an irrigation at FESW of 004 and refills the profile to 75% of FESW, this means that the next irrigation would be triggered after 78.8 mm of crop water use. The crop water on a daily basis is computed as the daily Class A pan evaporation times the ratio of crop water use to Class A pan evaporation as shown in Table 1004. 78 Table 10.4. Monthly irrigation guidelines for a low rainfall (1997-1998) and a high rainfall (1999-2000) growing season. Crops are ratoon crops started on 1 June, I August, 1 October and grown for 12 months where irrigation is scheduled at 004 FESW (135 mm soil water deficit) with a 30 day drying-off period. The number of irrigations and the ratio of crop water use to pan evaporation (pan factor ETlEp) are shown. Month 1997-98 Season of Irrigations 1 Jun Jun Jul Aug 2 Sep 3 Oct 4 Nov 4 3 0 nec 0 2 1 Jan 1 2 Feb 0 2 Mar 3 2 2 Apr I 2 2 May 0 2 2 Jun Jul Aug Sep 1999-00 Season oflrrigations 1997-98 Season Pan factor 3 3 0 1.18 1.29 0.34 0 0 0.93 1.43 1.22 0 0.81 1.08 1.31 0 0 0 0.80 0.88 1.40 0 0 0 0.75 0.80 0.99 0 0 0 0.67 0.69 0.76 0 0 0 0.46 0.70 0.74 1999-00 Season Pan factor 1.39 1.39 0.32 0.82 1.30 0.85 0.80 1.17 1.27 0.73 0.79 1.28 0.67 0.71 1.19 0.50 0.51 0.62 0.46 0.46 0.51 0.51 0.66 0.76 0.45 The frequency of irrigation is dependent on the evaporative demand, the occurrence of rainfall, and the crop water use. In general, the pan factors (ratio of crop water use to Class A pan evaporation) increases during early growth with canopy development towards a maximum value in excess of 1.0 and then decreases during later growth associated with the growth slowdown phenomenon (Table lOA). However, the actual number of irrigations each month of a given ratooning date varied with year associated with differences in the occurrence of rainfall and evaporative demand (Table lOA). In general terms, the data in Table 4 show that more frequent irrigation is required in the September to November period, but moderated according to the extent of canopy development and rainfall events. In practical terms, the data in Table lOA can be used as a guide to irrigation scheduling using the monthly pan factors for different ratooning dates and the observed Class A pan evaporation. When the pan factor times the Class A pan evaporation accumulates to 78.8 mm, then an irrigation would be triggered. However, these data only relate to one soil type (PESW of 225 and 75% refill at irrigation) and further research is required to test the wider applicability of these findings. 79 General Disclission Sustainable irrigation schedules need to maJ(1mlSe profitability and not necessarily productivity, and minimise water losses and accessions to the groundwater. This modelling analysis has shown that there is a trade-off between sucrose yield and irrigation water requirement when varying the frequency of irrigation. Where irrigation frequency is decreased from irrigation schedules that allow little soil water depletion to those that allow about twothirds of the plant extractable soil water (PESW) to be depleted before irrigation, there are considerable savings in the number of irrigations and thus labour requirements with relatively little penalty on yield. Water losses are also reduced in practice, as the irrigation application efficiency for furrow irrigated Cununurra Clay increases as the irrigation interval is lengthened (Muchow and Wood 1981). This will have important environmental benefits with reduced deep drainage and accessions to the groundwater, and reduced runoff. As the irrigation frequency is further reduced with irrigation after more than two-thirds of the PESW is depleted, dramatic reductions in sucrose yield and crop water use are predicted. The most profitable irrigation schedule will depend on the price of sucrose and the cost of irrigation relative to other growing costs. This research has made considerable progress in defining irrigation requirements and irrigation scheduling guidelines for the Ord Sugar Industry. Indeed, the current analysis is indicative of a lower irrigation requirement and less frequent irrigation schedules than the earlier analysis of Muchow and Keating (1998). The main reason for this difference is the observed growth slow-down phenomenon observed in two experimental crops in the Ord in 1998-99 and 1999-2000. More research is required to explain those observations and test their wider applicability as a basis for developing robust irrigation scheduling strategies for a profitable and sustainable Ord Sugar Industry. 80 11. IRRIGATION WATER APPLICATION EFFICIENCY Introduction The management of rising ground water levels in the ORIA is recognised as essential in ensuring the sustainability of irrigated agriculture in the region. Low irrigation water application efficiencies in the predominantly furrow irrigated bays are contributing to rising groundwater through runoff and deep drainage. This has become more evident with the changes in land-use in the ORIA with the re-introduction of sugar cane as a long season crop. The cause of tile rising water tables is often associated with deep drainage. Whether that occurs in the bays, tail drains, supply channels or in tile drains is not cIear. Rose et al. (1979) observed long term changes in salinity profiles under rice bays and furrow irrigated sugar cane/safflower bays over a 9 year period in the ORIA (1954 - 1963) indicating a drainage flux under the rice bays of 80 mm/year. Gordon and Gardner (1997) did some work based on chloride contents in several profiles and assuming a steady state, predictions of deep drainage Were made. The magnitude of the deep drainage varied with soil type, water salinity and applied amount of irrigation. The lowest predicted value was 13 mm/year and the highest was 69 mm/year for an irrigation salinity of 0.2 dS/m. Kinhill Engineering Consultants Ltd. (1999) in a feasibility study for Ord Stage 2 calculated annual accessions to groundwater using a water balance model (LEACHM). That varied from 56 mm/year to 119 mm/year depending on soil type and hydraulic conductivity associated with the soil profile. The maximum accession always occurred immediately after tile wet-season during March when the profile was completely saturated with very little deep drainage occurring for tile rest of the year. The latter has been frequently observed during the course of this project using permanently installed soil moisture measuring equipment. Banyard (1983) found significant drainage losses after some intensive studies by the Water and Rivers Commission through blocking-off major drains and supply channels. Measuring the infiltration it was found that the seepage losses in tile channels varied from 3 mm to 60 mm per day over the water surface and in the drains from 8 mm to 1000 mm per day. This equated to about 19 to 33 mm per year across the irrigable area in the ORIA from the channel infrastructure only. It appears that the rising water tables have several causes and inadequate on-farm application efficiency (AE) is one of tllem. Improving the AE by minimising the losses through the tail drain and deep drainage would therefore be a positive contribution to the management of the water tables. The assessments of opportunities to improve AE was aided by a surface irrigation model (SIRMOD) (Walker, 1993) which has been tested extensively both overseas and in Australia mainly in non-swelling soils. In the ORIA during the course of the project irrigation AE's were investigated for a number of furrow irrigated bays of sugar cane on Cununurra clay soils. This section presents data collected on those bays, the results from computer simulations using those data, the various approaches used in the simulations and examines possible options for improving water application efficiency. 81 Methodology Field measurements To quantify current irrigation practices and have data available to validate the computer simulation results several irrigation parameters such as inflow, outflow, soil moisture deficit at irrigation, deep drainage and advance rate needed to be measured. Inflow and Outflow A measure of the total water applied to the top of the bay (inflow) and amount lost by runoff at the bottom of the bay (outflow) were determined. Both flows were gauged in several furrows at the top and bottom of the bay in a similar way by means of paddle wheel flowmeters connected to a datalogger. The paddle wheel produce an electrical signal with every revolution which is recorded and stored in the memory of the logger and downloaded to a PC at regular intervals for further processing. The flowmeters were housed in rigid PVC pipes and installed in furrows prior to each irrigation see Figure 11.1. Each flowmeter and pipe combination required an individual calibration which was obtained using a commercial water meter from the WA Water Corporation, a bucket and a stop watch. I IEarthen dam or metal sheet Paddle wheel Data logger Water level Figure 11.1. Flow meter 'uo,"u,~u PVC pipe and placed in the furrow Soil Moishlre Changes in soil moisture content, i.e. Soil Moisture Deficit (SMD) were obtained from measurements using a neutron moisture meter probe (NP) before and after each irrigation event or on some occasions a capacitance probe or EnviroSCA~ (ES). The measurements were taken at depths of 20, 30, 40, 50 and 60 cm and then at 80, 100 and 120 cm. Prior to the NP measurements a standard count was obtained in a drum filled with water following the procedure according to Greacen, et at. (1965). The counts of the NP were divided by the standard count and the ratio related to a moisture content using a calibration curve obtained from extensive soil sampling. The position of the neutron moisture meter access tube had a significant effect on the deficit measured. On some occasions with the access tubes positioned in the middle of the beds, only a small increase in the moisture content following an irrigation was measured. This was particularly the case with deeper furrows. After several trials it was found that positioning the tube in the shoulder of the bed gave a more reliable estimate of the average SMD in the root zone. The measurements with the NP usually occurred about 1-2 days before and 1-2 days after irrigation for practical reasons. During those periods and during the irrigation evapotranspiration by the crop continues. This amount can be substantial under hot, dry and 82 windy conditions (and had to be estimated) but was not included in the calculation of the SMD. Deep drainage Initially the deep drainage component was estimated by difference from Equation (1) D=I-R-(DSM + Et) (I) where I the irrigation applied in mm, R the runoff in mm, DSM the change in soil moisture before and after irrigation in mm and Et the evapotranspiration occurring over the period between the NP measurements. It was found that in the low permeability soils (i.e. D expected to be small) of the aRIA this method was inaccurate. It was difficult to obtain a representative and accurate DSM as well as a reliable Et value. In addition an accurate I from the inflow and an R from the outflow measurements were also difficult to obtain with variations in magnitude (> 5 mm) between the individual furrows being similar to the expected deep drainage. Another way to measure the deep drainage employed in irrigation research and in this project is to compare the inflow with the outflow when both have reached the steady state (i.e. no change in the flow is detected). This state, particularly in the outflow is often not achieved in the low permeability soils of the aRIA because the duration of common irrigations does not allow for that. Twice irrigations were extended to 48 hours. On one occasion the profile was fully wetted up and the outflow became similar to the inflow (i.e. indicating no deep drainage). On the other occasion the outflow remained smaller than the inflow even though the profile was not fully wetted up. Under such conditions the difference between the inflow and the outflow is then not solely deep drainage but includes also a component of continuous wettingup of the beds. This was found to be the case particularly with deeper furrows. Advance rate, water depth and furrow shape Irrigation water advance rates during irrigation, which were used to calculate infiltration parameters, were obtained with resistor type sensors connected to a datalogger positioned at regular intervals down the furrow. The moment the water reached the resistor, the connection was shorted out across the resistor and the datalogger recorded a substantial drop in voltage which was clearly identifiable and could be used as a time stamp. The depth of water in the furrow was obtained in a similar way but rather than using one resistor, several (up to 20) resistors were positioned in series along an insulated rod. With an increase in water level, the total resistance of the series of resistors dropped as the resistors were shorted out one by one. The change in resistance was converted to a change in water depth, using a calibration equation. This was replaced by manual measurements during the irrigation at times when the equipment was not available. The furrow shape was determined manually prior to the irrigation and measured across the furrow at various positions. From the water depth and the furrow shape, the hydraulic radius (RR) defined as the ratio of the wetted area and perimeter in the furrow can be calculated. The RR is a major determining factor in the velocity of the water in the furrows. Modelling using SIRMOD SIRMOD model Modelling of surface irrigation has been carried out to assess opportunities for improving irrigation efficiency. The model, SIRMOD (Surface Irrigation Model) developed at the University of Utall (Walker, 1993) was nsed in this study. It describes the flow through the furrow whilst allowing for continuous losses due to infiltration. The infiltration or infiltration 83 properties of the soil are described using the Kostiakov-Lewis (K-L) empirical equation: 1= k t' + t fo where I is the cumulative depth of infiltration, k and a are fitted parameters, t = the time and fo represents the infiltration at very long elapsed times. Both parameters, k and a were determined from actual irrigation events using an optimisation program developed at the University of Southern Queensland, Toowoomba (McClymont, 1995). For this calcnlation the inflow, advance rate of the water down the furrow (distance + time) and the wetted area were torequired. The was determined from the difference in inflow and outflow when both had come to a steady state. Model testing In the initial stages of the project the objective of the simulations were to test the ability of the model to simulate actual irrigation events using infiltration properties of the bay derived from measurements during the irrigations. At a later stage several bays were monitored to test the model for its ability to simulated irrigation events when bay length and duration were altered using infiltration properties obtained from previous irrigations. The bays length and the duration were varied because those two parameters would not affect the infiltration properties of the soil. The infiltration properties are described empirically by the K-L equation and depend therefore on many factors that might affect the irrigation performance such as surface roughness, furrow shape, inflow, but also antecedent soil moisture. The bays length was altered by positioning outflow meters halfway the bay as well as at the end of the bay. Leaving certain furrows run longer altered the irrigation duration. Model Extension Instead of using infiltration properties for the different bays based on actual monitored irrigations it was investigated whether a general approach could be found to extend the use of SIRMOD to predict actual irrigations rather than simulating retrospectively. Three approaches were used: average parameters, Inflow-Outflow (IO) and the Instantaneous Crack-Fill (ICF) method and these will be described in the following sections. Average Parameters The use of individual infiltration separate parameters of each bay which depend on the growth stage of the crop, inflow, surface roughness, will produce very accurate simulations of the actual irrigations. This however requires elaborate and expensive measuring of irrigation parameters under many different scenarios. One approach to avoid the need for these measurements was to use the mean of the parameters obtained from a number of different irrigations in different bays across the ORIA. It was assessed whether one set of infiltration parameters (i.e. the mean) would be sufficient to describe the irrigations sufficiently accurately without any prior knowledge of the actual irrigations. Inflow-Outflow The approach of the Inflow-Outflow (IO) method is based on the concept that with ~ knowledge of the irrigations, i.e. inflow and the outflow, an estimate can be made of the total amount infiltrated, being the difference between the total inflow and outflow. With the information of the total infiltrated amount after for example 24 hours, an appropriate infiltration curve can be selected to simulate particular irrigations. The range of infiltration curves is illustrated in Figure 11.2. 84 mm infiltrated after 24 hrs. -<>-132 -D-95 --tr-73 -0-56 -<>-27 200.---------------------------------~" 180 +-________________~2~4~H~o~u~rs~________~~~--4 E160+_----------------~----_=~--------~ ~140 -1--------------7:7"=-----~-""-.y c o~120+_-------------~~--------~~~----~~ E100+-------~~~--_,~~~--~~~--~~ -'"~ 80 +-----~~--_r~Lf---~~~~--~~~~----__i ~ 60+-~~~~---~-£~~~-=~------------~~ :s 240o s °1~~~==~j o 500 1000 1500 2000 2500 3000 Irrigation duration (Minutes) Figure 11.2. Range of infiltration curves used in the Inflow-Outflow method Instantaneous-Crack Fill From ES observations in block B27 it was obvious that infiltration into the clay soil happened initially very fast, as can be seen in Figure 11.3. 140 827, 34 days since last irrigation 120 r E E ..; 100 c ~ 0 80 -.E'.", :.'':"".:. 60 40 ;c;:: 827, 14 days since last inigation If . .-.- - 855, ---~ 10 days since . ------ - l-a-s-t ir-ri-g-at-i-on J'"20 ..'.' o' 0 0:00 4:48 9:36 14:24 19: 12 0:00 Figure 11.3. Infiltration behaviour ofB27 at 34 and 14 days since last irrigation and B55 at 10 days since last irrigation. From Figure 11.3 it can be seen that in less than 4 hours the profile ofB27 is full. This is very common for the swelling and shrinking clay soils and has been described by various researchers such as Mitchell and Van Genuchten (1993) and Austin and Pendergast (1997) but not well documented in terms of real time infiltration measurements. This behaviour can be described as Instantaneuos Crack Fill (ICF) bearing in mind that it still took about 4 hours for the profile to fully wet up. The volume of the crack is solely drivell by the soil moisture deficit prior to irrigation, as illustrated in the figure by the difference in B27 between the 14 and 34 days since last irrigation. From the ES observation it was also 85 concluded that deep drainage during the irrigations was very small and probably could be ignored. The description of the infiltration became therefore a matter of applying water to the surface that rapidly i.e. in 4 hours filled the cracks with the deep drainage component during irrigation being ignored. The newer version of SIRMOD (V6-4-98) allowed for the infiltration description by including a crack fill component with the K-L equation which then becomes I = CF + k t' + t fo with CF as crack fill and the rest as in Equation 1. In this approach k and fo were equal to zero and CF solely a function of the soil moisture deficit at irrigation. There was no need therefore to obtain advance rate data to derive the infiltration parameters. This approach was satisfactory for several bays but not for all of them. Those bays did display a continuous infiltration during the irrigation and was far from instantanous. This behaviour is illustrated in Figure 4 by the infiltration at B55, where it took at least 14 hours to reach a maximum moisture content. This had to be reflected in the infiltration properties and SIRMOD was used employing the full K-L equation including a CF component with k, a and fo NOT equal to zero (non-ICF), which in essence is fairly similar to the Inflow-Outflow method. Results andDiscussion Field observations A total of 45 irrigations were observed. All bays were planted to sugar cane at the time but the cane was at different stages of development. The measurements were generally taken during the dry season (April - end of November) since the irrigations are fairly regular during that period. These included commercial (30) as well as experimental irrigations (15). On a number of the irrigations the position of the NP tube had been such that unreliable soil moisture deficits were obtained so that the AE could not be calculated. The median irrigation duration of the commercial irrigations was 24 hours, applied water was 124 mm, outflow 41 mm, irrigation AE 61 %, duration of the outflow 15 hours and the time to the tail draiu 10 hours. The results of all the irrigations are presented in Table 11.1. Wood et al. (1998) published a paper on benchmarking irrigation practices in the ORIA. Irrigation data such as frequency, number and duration of irrigations, time to reach the tail drain, duration of tail drain flow, water applied per irrigation and total water applied were obtained from a survey of sugar cane farmers. The median irrigation duration presented here is similar to that published. However the duration of the outflow is 8 hours longer than the survey found whilst the time observed for the water to reach the tail drain is 5 hours shorter. There is obviously a discrepancy between the measured times and the times assumed by the growers. Tail drain flow duration, even though it is significant in the overall irrigation AE, is not a parameter that irrigators use as an important criterion, but perhaps it should be. In the ORIA irrigation duration is often determined by fixed time intervals such as 12, 18 or 24 hours rather than a maximum duration of tail drain flow. The duration of outflow in this case is thus the result of a given irrigation duration. Surprising and difficult to explain however is the discrepancy between our observed median water applied per irrigation of 124 mm and the range found in the survey of 320-380 mm applied. A possible but unlikely explanation is that substantial losses occur between the Dethridge Wheel (water record of the farmer) and the furrows (water record of the research project). Ifthis were the case further work is required to reduce these losses. 86 Table 11.1. Summary of the data collected from commercial block irrigations. C = a commercial irrigation and NC a non-commercial irrigation Block B19 B27 Date Type Irrigation Deficit Duration (hrs) i (mm) 12105198 C 18 17108199 C 18 82 B27 31108199 C B27 14109199 C 12 18 48 58 B27 28109199 C B27 28109199 C B27 12110199 C B27 18109100 NC B27 12110100 C B2b 25108198 NC B2b 11110198 NC B2b 11110198 NC B2b 11110198 NC 24 24 18 48 18 11 6 6 6 65 125 71 66 71 67 5 8 8 B2b 7112199 NC B2b 8112199 NC B3b 24109197 NC B44 14110198 NC 8 10 15 18 15 33 54 B44 14110198 C 25 65 B51 1109198 NC 16.5 Dry B51 1109198 C 24 Dry B54 14109198 NC B54 14109198 C 23 20 Dry Dry B55 20103197 C B55 2104197 C 36 42 B55 11104197 C 25 B55 19104197 C 25 B55 28104197 C B55 7105197 C B55 21105197 C B55 31/05/97 C B55 9/06197 C B55 20/06/97 C B55 5/07197 C B55 23/07/97 C B55 24/04198 C B55 4/05/98 C B68 12108198 C B68 12108/98 C B69 20/04/99 NC B69 20/04/99 C 31 29 25 27 23 22 25 22 18.5 24 30 24 10 15 101 65 51 56 B69 19/05/99 NC B69 19/05/99 C B76 24/10/00 NC B86 20/08/98 NC B86 20/08/98 C 8 12 48 12 24 38 40 33 Dry Dry Median1r 24 Min* 12 Max'" 42 , JJetermmed from on y the commercia OlOCKS Ir';;:~~ 108 134 84 76 145 171 107 301 108 136 68 56 56 94 131 170 89 122 75 121 139 127 176 202 127 132 198 170 158 160 121 114 101 103 98 123 125 108 93 145 78 111 339 71 151 124 76 202 ~~~{' 40 24 26 4 70 41 39 225 33 47 33 28 16 43 64 46 45 60 36 60 46 22 50 27 34 64 45 43 48 42 30 43 24 56 37 32 12 53 76 29 48 199 41 107 41 4 107 Percent O~t~~~ from Inflow % 37 18 31 5 48 24 36 75 31 35 49 50 29 46 49 27 51 49 48 50 33 17 0 25 21 26 32 26 27 30 35 26 43 23 57 30 26 11 57 52 37 43 59 58 71 30 5 71 Application Efficiency (%) 61 57 76 45 73 66 22 66 49 7 14 14 16 25 61 53 81 60 55 39 49 36 10 61 36 81 Outflow (hrs) 14 6 7 2 10 16 10 10 5 4 4 4 8 7 8 14 24 12 20 13 15 18 30 15 28 24 20 24 13 7 9 14 7 8 10 22 15 2 30 Time,t~ T:~I Drain Hrs 10 10 6 10 10 6 8 8 4 2 2 2 4 3 8 5 5 7 7 13 8 20 17 14 12 10 11 12 21 21 7 7 7 7 4 4 10 4 21 87 SIRMOD simulations In irrigation modelling the two output parameters, advance rate and the outflow, are normally used to compare simulated with the actual irrigations. Using measured irrigation parameters Nineteen irrigations have been simulated using measured infiltration parameters from individual irrigations. A sample of these results is presented in Table 11.2. Table 11.2. A sample of simulation results using infiltration parameters derived from actual irrigation events. Bay Length Inflow Irrigation Water Adv. Adv. Runoff RunoffSim. (m) (l/sec) duration applied Meas. Sim. Meas. (mm) (hrs) (mm) (min) (min) (mm) B51 490 1.11 16.5 66 176 165 53 48 B68 1200 2.00 24.0 108 1338 1389 12 9 B2 180 1.12 11.0 135 296 330 46 46 B55 I 550 1.37 22.75 120 678 671 42 39 B55 2 550 lAO 22.3 115 747 700 30 28 Adv. - Advance tIme to end ofbay; Meas. and Slm.: Measured and Simulated respectIvely. The inflow sometimes varied during the course of the irrigation, whilst variability between furrows was sometimes considerable but overall there is a good agreement in Table 11.3 between measured and simulated advance rate and the runoff. The results of all nineteen simulations are depicted in Figure 11.4. Whilst good agreement between the measured and the simulated advance rate was achieved, the simulated runoff was generally slightly underestimated compared with the measured. .s 1400 E Ti 1200 ,gJ!l 1000 -E;;; 800 ~ .• " ' t·O< ,§ 600 ~uc 400 ~> 200 ".'.·ii "'C ....fIT a a 500 1000 .... j . a 1500 Advance time, measured, min 300 E 250 E -0200 J!l !! E150 ';; -..'. .,~ 100 - !E .•..0 50 ..-a '1'" ..1i'. •••• .. .' a 50 100 150 200 Outflow measured, mm .' •.'.' " b 250 300 Figure 11.4. Measured vs.simulated advance rates (a), Measured vs.simulated outflow (b). Simulations with variable bay length and duration Several irrigations were arranged to examine the response to different irrigation durations and lengtlls of furrows. The same infiltration parameters, a, k and fo as used for Table 11.1 were used in the simulation model to predict advance rate and runoff. The results of these simulations are presented in Table 11.3 and depicted in Figure 11.5, 88 Table 11.3. Simulation results for Bays 51 and B68. Block Length Inflow Duration Applied Adv. Adv. Sim. Runoff Runoff (m) (I/sec) (hrs) (mm) Meas. (min) Meas.(mm) Sim. (mm) (min) B51 240 1.11 16.5 153 176 165 108 99 490 1.11 16.5 75 402 435 36 26 240 1.23 24 245 176 152 169 155 490 1.23 24 120 402 427 60 47 B68 550 2.00 24 197 652 639 99 88 1000 2.00 24 108 1389 1338 12 9 550 1.85 30 227 652 695 151 III 1000 1.85 30 125 1389 1524 32 17 Adv. - Advance tnne to end of bay; Meas. and Snn.: Measured and sImulated respectIvely. 1800 1600 .E E.,1400 ~ ~ 1200 g~ 1000 !1 •"C 800 ";.•s; .Ii 600 400 15100 0 50 5 10 15 Measured advance time, hrs 20 y= 0.96x-1.18 ..R2 = 0.97 ' ,.• ..' •.' • ......... ~. •.' . ." .' b 50 100 150 200 250 300 Outflow measured, mm Figure 11.7. The simulated advance time compared to the measured time (a) and the simulated outflow compared to the measured outflow (b) 90 Using the measured inflow and the outflow and assuming an infiltration pattern, a good correlation was found between the simulated advance rate and the outflow and the actual measured time and outflow. The simulated advance rate tends to be underestimated for the longer durations without affecting the total outflow much. It is clear that with a little more information on the irrigations some good predictions of the irrigations can be made. Infiltration Behaviour and Crack-Fill The infiltration behaviour is described by the total amount of moisture stored in the soil as a function of time. Figures 11.8 and 11.1 0 illustrate that behaviour for various conditions encountered in several bays in the ORIA. As part of the study in several bays NP access tubes were installed at various positions down the furrow. Due to the position of the tube relative to the head ditch the inundation time at each position varied and could be determined. The change in soil moisture content at every position following irrigation was measured with the NP and plotted against the infiltrated amount at that point. •160 140 E E120 --5100 •0 ~ 80 • • •".e 60 ~ • -,--~ •~ 40 .s 20 827 • • 90 80 --E E 70 c 60 ~ 0 E 50 ~ ." 40 *30 ~ ""Ii: 20 .5 10 00 0 Inu1n0dation time,1(>rs 30 0 • ••••• 869 I 5 10 15 Inundation tim e, hours 20 Figure 11.8. The infiltrated amount at irrigation as a function of the duration of inundation for a number of irrigations at Bay 27 and Bay 69. As can be seen from Figure 11.8 no clear correlation was present between the inundation period and the amount infiltrated. The variation of the water applied was driven by the soil moisture stored in the profile prior to irrigation rather than the inundation time. The soil moisture stored in the profile determined the crack volume, as can be seen in Figure 11.9. 250 ,-------------------~--_, •§200 .c.... 827 5150 E a"l 100 ~ ;.;s: 50 0+----..,.----,---,----1 400 450 500 550 600 SM at irrigation, mm 80 ,--------------------=~~ E 70. 869 .E.... 60 . §o 50 E" 40 ]30 ~ 20 .s0;: 10 Oi-------,------r------~ 300 400 500 600 SM at irrigation, mm Figure 11.9. The depth of infiltrated amount at irrigation as a function of the soil moisture (SM) just before irrigation at Bay 27 and Bay 69. 91 As can be seen in Figure 11.10 some bays displayed a different infiltration behaviour. Rather than the total infiltrated amount being driven by the SM at irrigation it appeared to be driven by the duration of inundation. 70 •• 886 E60 E 7 -•540..,50 c y = 2.0781 x ~ 4.4423 m "w 30 f20 •.~s 10 O~-------r-------.------~ o 10 20 30 Inundation time, hrs 120 ~I .S100 E '•oE 80 ~ 60 868 y=2.6442x+33.676 R' = 0.626 •"!~!! 40 •~ 20 O+-----__- - - -________----~ o 10 20 30 40 Inundation time, hrs 70 E60 E ..,50 y= 2.9314x· 21.809 •c R2 = 0.6613 040 E •m ,,30 w ••~20 • •~10 O+--------r--------r-------~ o 10 20 30 Inundation tim e, hrs 90 80 E E 70 •E 60 o 50 E ; 40 ~ 30 .eb' 20 .E 10 I. ICentre o E44 ~10 Edge I y = 3.98Bx • 36.305 R2= DAi0S •• • O + - - - - -_ _- - - - - -_ _ _ _ _ _ _ _~ o 10 20 30 Inundation tim e, hrs Figure 11.10. Infiltrated amount as a function of the inundation period of four bays (B86, B68, B51 and B44). The duration of inundation had a great effect on the infiltrated amount. In all of these bays the tubes had been placed in the centre of the bed hence the NP readings only reflected the moisture changes in the centre of the bed. These changes often did not correspond with the total amount infiltrated as calculated from the difference in the inflow and the outflow of the bay. It was therefore concluded that the soil closer to the furrow had to be wetter than the beds to compensate for the dryness of the beds. The difference in the stored amount of water in the edge and the centre of the bed, as shown in Figure 11.10 for B44, confirms that trend. A major difference between B27 and B69 and B86, B68, B51 and B44 is the shape of the furrow. The furrow profile of B27 and B69 was more shallow and had a distinct flat-base profile whilst the other four blocks had deeper and more distinct 'V-shaped profiles. Figure 11.11 illustrates the various profiles. The effect of furrow profile on irrigation performance is also illustrated by the following. Bay 27 changed from a shallow profile in 1999 to a deeper furrow in 2000, see Figure 11.11 (B27, 18/09/00). This was done because the grower was not happy with the furrow profile and how the bay drained during the wet season. This change in furrow shape/depth affected the behaviour of the subsequent irrigation. From a slow moving rapidly infiltrating furrow it changed to a fast moving slowly infiltrating furrow. The irrigation in September 2000 (18/09/00) with a similar inflow rate and at a similar soil water deficit as an irrigation in September 1999 took 5 hours less to advance to the end of the bay due to the change in furrow profile/depth. The change in furrow profile had a large effect on the performance of the 'irrigation. The deeper 'V-shaped furrows took longer to wet up, they reached the tail drain faster, the runoff losses were higher and therefore resulted in a lower AB. 92 Distance across the furrow, cm 50 100 150 10E +----\--\,+-""'=---tJ----'c---~~-_+_ +-___S" .3 15 ~~~~~~;;::~~L~ 20+-----------~~ -tr-868 -X-844 -:1:-851 --:I'''---,,"~''''--------1_ 854 -82b 25 -'-_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _-j-+-0-8-827217MI9l00 --{i}- 855 Figure 11.11. Furrow profiles of the bays as iudicated in the legend. The profiles were taken just prior to the irrigations. The reasou for this difference appears to be in the hydraulic radius (HR) of the furrow. The HR is defined as the ratio of the wetted area CA) and the wetted perimeter (P) during the irrigation. There is thus a theoretical relationship between HR and the water velocity in a channel, according to the Chezy formula which is defined as: v =C (HR X 5),,112 (2) where V is the velocity in meters per second, C is a coefficient dependent on the roughness of the channel (Manning's n), HR and the slope s. So with an increase in HR, an increase in the velocity is expected. In Table 11.5 several furrow properties as well as other dimensions ofthe flow are presented such as A, P, HR and water depth. Table 11.5. Furrow and hydraulic flow dimensions, furrow inflow, furrow length, time to reach themt'l dra'mand tlle advance rate m. tlle furrow Dor varIO. US bays. Bay Furrow A depth (cm2) Water P HR Inflow Furrow Time to tail Advance depth (cm) (em) (Vsec) length drain rate (cm) (cm) (m) (min.) (m/min) 44 22.1 213 5.5 61.1 3.49 1.44 580 500 1.16 51 21.7 167 5.8 51.3 3.26 1.11 450 402 1.12 27 9 124 3.1 60.3 2.06 0.95 300 840 0.36 54 13.9 339 7.8 81.6 4.16 1.05 350 555 0.63 86 21.3 516 12 74.5 6.93 1.2 430 240 1.79 68 22 251 8.5 53.5 4.71 1.8 1000 1338 0.75 69 14.1 268 5.5 81.1 3.31 1.2 260 446 0.58 2B 13.9 228 3.9 91 2.52 1.07 170 296 0.57 27, deep 21.6 299 6 91 3.28 0.8 300 480 0.63 76 15 414 9.2 83 5 1.2 400 529 0.76 69 14.1 234 9.1 71 3.3 1.2 260 453 0.57 Graphically displaying the data from Table 11.5, as is done in Figure 11.12, generated some useful relationships. 93 8 7 -, E "-. "~ 4 ~ •3 ~2 0 0 y"'0.4651x +0.59 R~ '" 0.8273 • • , 10 I-\'dnulicradlul (em) A " 2.0 ;G1.8 .E 1.6 E ;-1.4 ~ 1.2 ~ 1.0 ~ 0.8 : 0,6 ~O.4 ~ 0.2 0.0 0 y: 0,2235x - 0.0434 R2" 0.5546 .- • 24• Hydra"ullc radius (em) B 2.0 :S'1.8 y=O.0584x-O.1918 • .€ 1.6 E -;1.4 R2 .. 0.4557 'E 1.2 / -i81.0 O.S co 0.6 Q •I! 0.4 <0.2 c 0.0 1--~-~--~-_--l • "o 10 15 20 Furrow depth (em) Figure 11.12. The water depth in the furrow as a function of the hydraulic radius (A), the average advance rate in the furrow as a function of the hydraulic radius (B) and the average advance rate as a function ofthe furrow depth (C). There is clear relationship between the measured water depth in the furrow and HR. There is also some relationship (R2 = 0.55) between HR and the advance rate in the furrow even though the relationship is not as good because the flow in the furrow is not solely dependent on HR but also on C and the slope. A similar relationship exists between the depth of the furrow and advance rate. These relationships provide an explanation for the measured dependency of the advance 'rate on the shape/depth of the furrow. With a change in furrow shape the water depth in the furrow is increased, which in turn increased the velocity of the water in the furrow. It is difficult to separate the shape from the depth ofthe furrow because they are confounded in the commercial irrigation furrows and thus have been used in conjunction. From an irrigation application efficiency point of view is would be more advantageous to have shallower and broader furrows in which the water flows more slowly with more opportunity for crack fill and less tail water runoff. From a drainage point of view in the wet season the deeper furrows might be preferred even though with properly laser levelled bays and clean tail drains this issue might be oflesser importance. In the sandy loam soils of the Burdekin Delta (Qld) where the effect of furrow shape was investigated by the BSES (Holden et at, 1998), it was found that the deeper 'V'-shaped furrows had a better AE than the 'V'-shaped furrows. Where lateral infiltration into the beds is not a problem, the deeper 'V'-shaped furrows limited the overall opportunity time for infiltration because water in these furrows reached the tail drain faster than the 'V'-shaped furrows. In the Burdekin Delta deep drainage losses are significant and by chauging furrow shape from a 'U'shape to a 'V'-shape these can be substantially reduced. This issue does not apply to the ORIA, at least not in the bays we have investigated, because the deep drainage compouent tended to be much smaller. Modelling Crack Fill As was shown before, SIRMOD can be used to simulate various irrigations accurately both in terms of predicted advance times and runoff, two parameters often used in the assessment of 94 irrigation simulations. The third parameter is the amount infiltrated and stored in the root zone during the irrigation. On all occasions soil moisture changes detected with the NP should reflect the calculated moisture stored in the root zone. The application of generalised infiltration parameters did not provide good estimates of the advance rate and the runoff. The modelling approach was then changed and an allowance made for infiltration behaviour typical for clay soils whilst distinguishing between 'U'-shaped and 'V'-shaped furrows. The approach with the use of SIRMOD has been two fold. Firstly the instantaneous crack fill (IeF) has been used for those bays that have 'D'-shaped furrows and secondly the non-IeF method was employed for bays with 'V'-shaped furrows. Simulations using the IeF for the 'U'shaped furrows gave the following results, presented in Table 11.6. Table 11.6. Simulated and measured outflow, advance time and infiltrated amounts (1nfiltr) using the IeF method with negligible deep drainage. Bay Date Outflow Outflow Sim (m3) Meas_(m3) 27 16/08 12 12 12110 22 21 31/08 18 ~8/09 35 13 35 28/09 22 21 2B 125108 14 14 7112 17 12 8/12 20.5 18 54 14/09 29 29 Sim - SllUulated, Meas. - Measured Advance Advance Infiltr. Infiltr. sim{min) Meas (min) Sim (mm) Meas (mm' 974 960 111 111 757 640 71 80 465 420 47 55 790 630 75 80 1121 960 128 128 480 300 96 96 219 164 35 40 302 229 58 58 954 991 94 94 Good agreement was achieved in B27 and B54 between all three measured and simulated parameters ONLY using the crack fill component, based on the soil moisture deficit as a variable. In 2b the agreement was good in the outflow but not in the advance rate. It appears that t11is bay might be subject to some deep drainage. When SIRMOD is applied to bays with the 'V'-shaped furrows, such as B44, it was found that the 1eF approach is no longer applicable. A more complex infiltration description is then necessary including a component, the base intake, fo, which resembles the sorption into the beds as a long term infiltration component as well as the transmission component of moisture passing below the root zone as deep drainage. Little is known about the infiltration behaviour of the bays with distinct 'V'-shaped profiles when irrigated at larger deficits but it is reasoned that the crack fill component will become more pronounced, as illustrated in Figure 11.13 using a different deficit (Me) at irrigation. The constant infiltration rate later during the irrigation which resembled the sorption into the beds, is 60mml24hr = 2.5 mmlhr which is close to the average slope (2.65 mm/hr) of the regression lines fitted through the observation points in Figure 11.10. 95 0.14 . , - - - - - - - - - - - - - - - M S : I 0.12 120mm iEro: O. 1 I 100mm I 9~ ;;: 0.08 7o~m 1: ;o;: 0.06 53mIm I .c: 15. 0.04 tt,;.:::::==::;;:~-====::;;;;;--~---_j 8 0.02 O~-------_.--------_.--------~ o 500 Time, min 1000 1500 Figure 11.13. Infiltration patterns for B44, a block with a deep 'V'-shaped furrow profile for various levels of Moisture Stored (MS) during the irrigation. CF indicates the level of crack fill. Using the infiltration curve for a MS of 53 mm, based on the deficit at irrigation needing to be replenished, the following results presented in Table 11.7 were achieved simulating an irrigation in B44. Table 11.7. Simulation results using SIRMOD on B44 for different bay lengths and irrigation durations. Bay length Water applied Advance Advance Outflow Outflow and duration (mm) time (min) time (min) (mm) Meas (mm) Sim Meas. Sim 300m, 181m 174 195 187 120 118 300m, 25hrs 242 195 187 183 174 580m, 181m 90 500 446 45 36 580m, 25 hrs 125 500 446 60 59 Infiltration (mm) at different Meas. Sim, position in the bay: mm mm Top, 18/zrs 53 54 Centre, 18 hrs 54 53 Bottom, 18 hrs filled 48 Top, 251m 63 66 Centre, 25 hrs 67 65 Bottom, 25 hrs filled 60 Meas. ~ Measured, Slm. ~ Sunulated. FIlled - filled wIth water backed up the furrows and filled the NP tubes. It can be seen that the simulated advance time, outflow and the infiltrated amount compared very well with the measured even though the simulated outflows at 580 m and 18 hrs were slightly under-predicted. From these results it has been concluded that the current approach using the ICF and the nonICF method, would be suitable to develop look-up tables. The non-ICF is in essence similar to 96 the IO-method described earlier. Both describe the infiltration into the soil with a continuous infiltration component later in the irrigation. The non-rCF method still assumes a crack-fill component which is filled rapidly, followed by a continuous infiltration rate of about 2.5 mmfhr. The crack-fill component is determined by the deficit at irrigation. The IO-method requires knowledge of the total amount infiltrated and on that basis an infiltration curve is selected which looks very similar to the non-rCF infiltration curve. Look-up Tables There are several ways to improve the AB. The AB can be improved by changing the inflow rate, the bay length, the irrigation duration, the soil moisture deficit at irrigation, and so on. Look-up tables can be used to establish relationships between a number of variables that can be changed to improve the AB. Instantaneous Crack Fill (ICF) approach applied to U-shaped furrows Using the ICF approach the modelling of the irrigation becomes very straightforward because the infiltration properties are driven by the soil water deficit at irrigation which can be easily determined with in-field measurements or simulated with crop growth models. Irrigations were simulated for a different CF at irrigation and presented in Table 11.8. The soils at the various bays have been described in terms of Drained Upper Limit (DUL), Lower Limit (LL) and Field Capacity (FC), information often required by crop growth models. Table 11.S. Lookup table for Blocks 27 and 2B for irrigations of a single duration and different CF. Duration CF AE Runoff Runoff (mm) (%) (m3) IAppl B27 24 200 nr nr nr DUL=808mm 24 180 100 0 0 LL=540mm 24 150 83 15 0.15 FC=783 mm 24 III 62 34 0.37 24 100 55 40 0.44 24 50 27 65 0.72 28 180 84 16 0.16 B2B 10 150 nr nr nr !oUL=771 mm 10 130 100 0 0 LL=548 mm 10 100 76 8.9 0.23 FC=709 mm 10 80 61 14.6 0.39 10 60 46 20.4 0.54 DUL =Dramed 10 Upper LI.m.It; 40 LL 30 =Lower . .26.1 Lumt; FC 0.69 -FIeld Capacity achieved after normal irrigation, CF =Crack Fill at irrigation; AE = Application Efficiency; Ru = Runoff; RU/AppJ = Fraction Runoff to Water Applied, nr = not applicable because the water did not reach the end ofthe bay. The use ofICF is a good approximation of reality but an approximation nonetheless. As can be seen from Figure 11.2 it takes generally a little longer (ie. up to 4 hours) before the profile is saturated. It is therefore not feasible to apply exactly the same amount of water as the volume of cracks. The water needs to run for an additional 4 hours before the profile is saturated. This is illustrated by the 28 hr irrigation in Table I 1.8 during which runoff losses occur and a reduction in the AE from 100 % to 84 %. A more extended look-up table with variable bay lengths, irrigation durations and inflow is included in this report as Table 11.9. 97 Table 11.9. Look-up table for a flat base shallow furrow, characterised by an instantaneous crack fill with varying inflow rates, lengths of bay and duration of irrigation. Values in the matrix are the Application Efficiencies (AE) Zero deep drainage is assumed for this table and an additional irrigation duration of 4 hours to allow for a slower than instantaneous crack fiIl such as observed with the Enviroscan in B27 is included in the duration. 1 I1sec 1.5 I1sec 2I1sec CF-50mm Length, m lOhrs 18 hrs 24hrs 32 hrs Len!rth, m 10hrs 18 hrs 24hrs 32 hrs LeMth, m lOhrs 18 hrs 24 hrs 32 hrs IESW= 193 mm 200 50 28 21 16 200 33 19 14 10 200 25 14 10 8 Refill = 97% 400 ---- 56 42 31 400 67 37 28 21 400 50 28 21 16 IFC=783 mm IDUL= 808 mm U=540mm 600 800 1000 ---- ---- 63 47 ---- ---- ---- 63 ------- ---- 78 600 ---- 56 800 ---- 74 1000 ---- ---- 42 31 56 42 69 52 600 ---- 42 31 23 800 ---- 56 42 31 1000 --- 69 52 39 CF -IOOmm \ESW= 143 mm lRefiIl = 97% IFc = 783 mm IDUL= 808mm LL=540mm Length, m 200 400 600 800 1000 10hrs ------- ---- ---- ---- 1 I1sec 18 hrs 24hrs 32hrs 56 42 31 ---- 83 63 ---- ---- ------- ---- ---- ---- ---- ---- Len~h,m 200 400 600 800 1000 10hrs --- ---- ---- ---- ---- 1.5 I1sec 18 hrs 37 74 ---- ---- ---- 24 hrs 32hrs 28 21 56 42 83 63 ---- 83 ---- ---- Length, m 200 400 600 800 1000 2 I1sec lOhrs 18 hrs 24hrs 32 hrs 50 28 21 16 ---- 56 42 31 ---- ---- 63 47 ---- ---- 83 63 ---- ---- ---- 78 1 I1sec 1.5 I/sec 2I1sec ~F-150mm Length, m 10hrs 18 hrs 24hrs 32 hrs Length,m lOhrs 18 hrs 24hrs 32hrs Length, m lOhrs 18 hrs 24hrs 32 hrs ~SW=93mm 200 ---- 83 63 47 200 ---- 56 42 31 200 ---- 42 31 23 RefiIl = 97% 400 ---- ---- ---- ---- 400 ---- ---- 83 63 400 ---- ---- 63 47 iFc = 783 mm 600 ---- ---- ---- ---- 600 ---- ---- ---- ---- 600 ---- ---- ---- 70 IDUL= 808mm 800 ---- ---- ---- ---- 800 ---- ---- ---- ---- 800 ---- ---- ---- ---- II::.L= 540 mm 1000 -----_-.--- _ .. _ - ---- ---- 1000 ---- ---- ---- ---- 1000 ---- ---- ---- ---- DUL =Drained Upper Limit; LL =Lower Limit; FC =Field Capacity achieved after normal irrigation, PESW =Plant Extractable Soil Water = DUL-LL; CF =Crack Fill at irrigation; ESW =Extractable Soil Water still in profile at irrigation = CF - LL; AE = Application Efficiency = Depth Stored in Profile/Depth Applied as irrigation; RF =RefiIl = (CF + ESW)IPESW. 98 It should be pointed out that at no stage is the DUL achieved during the irrigation. The DUL is only obtained after prolonged ponding with enough hydraulic head for a sufficiently long time to wet up the deeper soil layers. This is never achieved during the irrigations, not even during the wet season. It was found that only moisture profiles in the tail drains and the supply channels will approach the DUL. They are probably also the places where most of the deep drainage occurs. Non-Instantaneous Crack Fill Approach In a similar manner with different infiltration functions the following look-up Table 11.1 0 was established for B44, a bay that has distinct 'V'-shaped furrows. Table 11.10. Look-up table for B44 which has a distinct 'V'-shaped furrow profile. B44 Duration, MS, AE,% RF,% brs mm 18 50 50 90 DUL=1111 mm 24 50 42 100 LL=872mm 24 70 55 94 FC=985mm 32 70 43 100 24 100 77 92 DUL -Dramed 32 Upper 100 LImIt; LL 68 -Lower L.m.1ut0;0 FC =Field Capacity achieved after normal irrigation; MS = Moisture stored during irrigation; AE =Application Efficiency; RF = Refill percentage As shown in the table the AE increases with MS for a given duration. More water is stored in the soil profile and less is lost as runoff. The 24 hrs at MS = 100mm results in AE of 77 % but the Refill (RF) is 92 %, indicating that the 100 mm is not applied across the bay evenly and the bottom section is under-irrigated. It should therefore be irrigated longer to achieve a RF of 100 % which results in a drop in the AE to 68 %. An alternative approach to the V-shaped furrows is to use smaller siphons to reduce the inflow and therefore reducing the runoff losses. A smaller inflow will however reduce the water depth in the furrow which might increase the time it takes to wet-up the beds. A more extended look-up table is provided as Table lUI for B44, a bay representing bays with deep V shaped furrows for variable bay lengths, irrigation durations and variable inflow rates. 99 Table 11.11. Look-up table for a deep V-shaped furrow, characterised by a slow infiltration process with varying inflow rates, lengths of bay and duration of irrigation. Values in the matrix are the Application Efficiencies (AE) Zero deep drainage is assumed for this tabl -. Il1see I.Sllsee 211sec ~SM 50 Length (m) 18 hrs 24hrs 32 hrs Length (m) 18 hrs 24 hrs 32hrs Length (m) 18 hrs 24hrs 32 hrs ~SW 63 200 ---- 21 15 200 ---- 14 10 200 13 10 8 J1C 985 400 ---- 42 43 400 --.- 27 21 400 ---- 20 15 DUL 1111 600 ---- 62 47 600 --.- 42 31 600 ---- 31 23 LL 872 800 ---- ---- 62 800 ---- 55 41 800 ---- 40 31 PESW 239 1000 .--- ---- 77 1000 ---- 69 52 1000 --.- 52 39 Il1see 1.slIsee 211see dSM 70 Length (m) 18 hrs 24 hrs 32 hrs Length (m) 18 hrs 24 hrs 32hrs Length (m) 18 hrs 24 hrs 32 hrs ESW 43 200 ---- 28 21 200 .--- 19 14 200 ---- 14 10 400 ---- ---. 43 400 ---- 37 29 400 ---- 29 21 600 ---- ---- 65 600 ---- 54 43 600 .--- 42 32 800 ---- ---- --.1000 ---. ---- --.- 800 ---- -.-- 58 1000 ---- ---- 72 800 ---- .--- 43 1000 .--. ---- 55 Il1see l.slIsee 211see ~SM 100 Length (m) 18 hrs 24hrs 32 hrs Length (m) 18 hrs 24hrs 32 hrs Length (m) 18 hrs 24 hrs 32 hrs ESW 13 200 ---- --.- 31 200 -.-- 27 20 200 ---- ---- 15 400 ---- --.- 60 400 -.-- 54 41 400 ---- ---- 31 600 ---- ---- ---- 600 ---- --.- 60 600 ---- -.-- 46 800 ---- --.- .--. 800 ---- --.- ---- 800 ---- --.- 60 1000 ---- --.- .--. 1000 ---- ---- ---- 1000 ---- --.- ---- Il1see 1.5l1sec 211see I ~SM 120 Length (m) 18 hrs 24hrs 32hrs Length (m) 18 hrs 24 hrs 32 hrs Length (m) 18 hrs 24hrs 32 hrs I ~SW -7 200 ---- .--- 37 200 --.- 32 25 200 ---- 24 19 400 ---. ---- 72 400 --.- ---. 50 400 ---- 47 37 600 ---- ---- ---. 600 --.- ---- 72 600 ---- ---. 56 800 ---- ---- ---- 800 --.- .--- _1001000 .--- ---- ---. L 0__ --.- ---. ----- dSM = Change in soil moisture following irrigation. ----: indicates an in-complete filling ofthe profile ---- ---- ---- 800 --.1000 ---- ----- ---- -- - - - ---- ---- .--- ---. 100 COllclusiolls The irrigation application efficiencies of irrigated bays of sugar cane were determined through direct measurement, with data obtained used in simulation modelling to predict optimum duration of irrigation and bay length. Measured runoff losses observed ranged from 5 to 71 % of water applied. The median of the measured commercial irrigations gave similar results as a grower survey in terms of irrigation duration and applied water but a longer duration of tail water flow and a shorter time for water to reach the tail drain. Deep drainage in most bays under irrigation that have been investigated by us appeared to be minimal and has been neglected in the modelling of irrigations using the ICF approach. Except for some bays it is assumed that deep drainage predominantly occurs during or shortly after the wet season. A time when little irrigation is used. The use of one set of average infiltration parameters did not simulate the irrigations accurately. Some more information in terms of infiltration behaviour using the InflowOutflow method is required to improve the simulation results. With some information on tile inflow and the outflow during the irrigations, good predictions can be made ofthe irrigation. In some bays in the ORIA the Instantaneous Crack-Fill method provides a good approximation of the infiltration behaviour. These bays have without exception 'U'-shaped furrow profiles. Other bays in the ORIA with deep 'V'-shaped furrow profiles were found to have a different infiltration pattern, characterised by long and slow infiltration. The simulation of the irrigation can be done successfully using tile ICF approach or a standard infiltration (non-ICF) with a base infiltration ternl depending on the shape of the furrow profiles but both as a function of the deficit at irrigation. The base infiltration term would reflect the sorptivity into tile beds and some component of deep drainage. The AE improves considerably with an increase in the deficit at irrigation both for bays Witll 'U'-shaped and with 'V'-shaped furrows. Given the lack of or the very small deep drainage component the best improvement in AE is achieved by minimising the runoff losses tllfough tile tail drain by extending the irrigation interval (i.e. increase the deficit at irrigation) by reducing the inflow and/or changing tile furrow shape. Accessions to the groundwater are facilitated by prolonged ponding in the tail drain and tile supply channels given the saturated profiles and the large hydraulic head in these features. Reduction of tail water runoff and the rapid removal of water from tile tail drains is therefore essential. 101 12. IRRIGATION GUIDELINES At the beginning of the project the growers were largely 'experimenting' with different irrigation scheduling methods. One of the key outcomes of this project was to examine the water requirements of sugarcane and develop irrigation scheduling guidelines for Ord growers. Many different methods of developing irrigation guidelines have been developed and introduced to industry over the years. Many of these various methods were examined for suitability for use in the Ord. A workshop to present the results of the irrigation work was held in early 2001. All data available was presented to those present. At this workshop a decision was taken, after much discussion and consideration, to develop a table of daily intervals as a guide to scheduling irrigation of sugarcane at a range of different harvest dates. Developmellt ofirrigatioll schedulillg guidelilles Recent research has provided the following four key pieces of information that have been used to develop the guidelines for use by farmers for scheduling furrow irrigation of sugarcane in the ORIA. 1) Plant available water holding capacity (PAWC), which is a characteristic of soil hydraulic properties and rooting depth, was measured in situ for three major soil type variations of Cununurra clay in the ORIA (Muchow et al., 2001). These soils are typically used for sugarcane production. 2) It was noticed that irrigation did not fully wet up the soil because of surface sealing after swelling of the 2: I lattice clays prevalent in these soils. The amount of water in the profile after irrigation ranged from 75 to 85% ofPAWC. 3) Irrigation trials indicated that cane yield was not affected until the crop had extracted about 50% ofPAWC. From this knowledge, farmers selected two target deficits, one at 50% ofPAWC, the other at 40% for irrigation scheduling. 4) The amount of water used by the crop on a daily basis was estimated by the APSIMSugarcane model. This model has been found to represent the growth of sugarcane under a wide range of conditions (Keating et at., 1999). However, under Ord conditions, radiation use efficiency and hence crop water use had to be reduced after the crop had reached a certain biomass threshold (30 tlha or about 80 tlha of cane) in order to match simulated and measured crop water use and yield. Estimates of daily crop water use were not allowed to exceed 1.25*ETa based on BREB research in the Burdekin and in Swaziland (McGlinchey and Inman-Bamber, 2002). Information on PAWC for varying soil types, percentage of profile filling with irrigation and target deficit was used to generate further information on water available for use by the crop between irrigations (Table 12.1). TIlls was then used in the APSIM model in association with long term climatic data from the Frank Wise Institute to provide a daily interval on which to re-irrigate cane at different stages of growth (Table 12.2). The system has an accuracy of (+12 days) which is more than adequate given that the sensitivity of the flood irrigation system is also about 2 days. This daily interval method was also selected with the knowledge that sugarcane, to a large extent is not responsive to irrigation practice, except under extreme irrigation schedules. Given that the sugarcane was unlikely to suffer long term damage with a system accurate to a within +1- 2 days and the irrigation system itself was also inherently inaccurate, it was difficult to justify increasing the accuracy of the scheduling system. 102 Table 12.1. The amount of water available for plant use for three ORIA Cununurra clay soil types with varying PAWC, assuming 80% refill with irrigation and with target deficit of 50% PAWC. Soil Type I 2 3 PAWC (mm) 169 190 225 Plant available water after irrigation with 80% refill (PAWC*80%) l35mm l52mm l80mm Target soil water deficit (PAWC*50%) 85mm 95mm 113mm Water available with 80% refill and 50% target deficit (PAWC*80%) - (PAWC*50%) 50mm 57mm 67mm When the results of the APSIM simulations were first presented to growers, best-bet intervals were provided for different rainfall scenarios. Growers however preferred to account for rainfall themselves and adjust irrigation intervals accordingly. Hence guidelines provided (Table 12.2) use simulations developed assuming no rainfall. Guidelines developed indicate that crops harvested on I August should receive an initial irrigation soon after harvest, with the second irrigation occurring 40-60 days later. However, these early irrigations are often influenced by individual farmer management practices, such as fertiliser application and herbicide incorporation. Scheduling for October through to January would require irrigation to occur every ten days, then thirteen days for February and March, and so on until July when the crop would begin the dry down period (DD) prior to harvest, with irrigation intervals adjusted for rainfall as necessary, throughout. These guidelines have provided an effective extension tool to quickly update irrigation recommendations based on current research work being conducted in the Ord. Growers have used the tables as a means to modifY irrigation scheduling in conjunction with their own experience in cane irrigation. As familiarity with scheduling principles increases, it is feasible that some growers will move to a more precise scheduling method, such as the use of crop factors and the water balance technique. Closer attention to matching crop water demand and supply will also be driven to a large extent by the likely increase in the cost of water and by requirements for more efficient use of water and improved environmental management. 103 Table 12.2. Irrigation scheduling guidelines for sugarcane for three harvest dates and three soil water deficits based on long tenn climate data (a) Conservative Schedule - to remove 50% of plant available water (FESW =0.5) Month ofyear Harvest date JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV 1 JUNE 10 11 11 12 DD 12 10 9 9 1 AUGUST 1 OCTOBER 9 9 10 12 12 15 19 DD 10 11 12 14 18 18 15 DD 9 9 (b) Optimal Schedule - to remove 60% of plant available water (FESW = 0.6) Month ofyear Harvest date JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV 1 JUNE 13 13 14 16 DD 14 13 11 12 1 AUGUST 12 13 14 16 20 24 DD 1 OCTOBER 11 13 14 15 18 23 24 20 _ -_- - - - L-________ - - - _.. - - ... - - - - - .. - - -~ - (c) Dry schedule - to remove 70% of plant available water (FESW = 0.3) - (85% refill) DD 11 11 Month ofyear Harvest date JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV 1 JUNE 17 17 18 21 DD (78) 18 16 16 1 AUGUST 16 17 18 21 26 32 DD (64) 1 OCTOBER 16 16 18 21 25 30 34 28 - Assumptions: - Refill of the profile is assumed to be 80% after each irrigatIOn - No rainfall - Schedules based on averages of three soil characterisation sites (Innes, Lerch, Frank Wise) DD 16 DEC 9 9 9 DEC 12 12 11 DEC 17 16 16 104 13. SECOND SURVEY OF IRRIGATION PRACTICES A second survey of irrigation practices was conducted in late 2000 and early 200 I. The survey focussed only on irrigation management, unlike the initial survey in 1995-96 which included fertilising practices and other farm management inputs. The main changes in irrigation practices that occurred in the 5 years between surveys were: • A slight increase in the time period between irrigations. • A reduction in the number of irrigations per year for some growers and little change for others. • Significant reductions in the total amount of irrigation water applied by a few growers but little change for the majority of growers. • A reduction in the duration oftail drain flow for most growers. The survey was conducted prior to the release of the best practice irrigation guidelines at the April 2001 growers' workshop. Whilst the growers had been kept well informed of the results of the ongoing irrigation research, it was not until early 2001 that clear and unambiguous guidelines were released. Consequently it is not surprising that only a few growers had modified their practices before their release. We are aware of more substantial changes in irrigation practice having occurred in tlle period after the growers' workshop. In particular, growers have adopted longer periods between irrigations, fewer irrigations per year and less total water applied per year or per crop. A further survey of irrigation practices should be conducted towards the end of the follow-up irrigation project (CSE007) in 2005. 105 14. GENERAL DISCUSSION AND CONCLUSIONS Research in this project has resulted in a number of significant and unexpected findings and has raised some important questions. In alI of the trials conducted there were no significant effects of drying-off on cane yield, Pol or Pol yield. This was a surprising result given the long drying-off periods of up to 80 days in some experiments. The reasons for the lack of response are not clear but it appears likely that at the time the drying-off treatments were imposed, cane yields were being limited by some unknown factor and that rates of crop water use at this time were relatively low. Based on the results from seven separate trials, drying-off does not appear to be an appropriate strategy in the Ord for increasing sugar concentration (Pol) in cane stalks. This is in contrast to the enhancement of sugar concentration or CCS often observed under Burdekin conditions, as described by Robertson et af (1999a). Consequently it is concluded that the drying-off strategy should only be used in the Ord for improving the trafficability of fields prior to mechanical harvesting and needs to be tailored to different soil types and to crop water use. Similarly to the drying off trials, there was little yield or Pol response to the different irrigation treatments in the schednling trials even though one treatment was fairly extreme in each trial. The only significant effect recorded at commercial harvest was a lower cane yield in the 180 mm pan evaporation treatment in the 1998-99 crop. The most severe water deficit treatment in the 1999-2000 crop also had a lower cane yield, Pol and Pol yield than other treatments although they were not significantly different. In general the yield reductions associated with extending the irrigation interval were not great when compared with the potential savings in water, labour costs and accessions to groundwater. However the two seasons when tile trials were conducted (1998-99 and 1999-2000) had rainfalI totals of 1145 mm and 1464 mm respectively which are welI above the long term annual rainfall of 786 mm. It is concluded that less freqnent irrigation strategies warrant further investigation particnlarly in the period following the wet season but the response to different irrigation schednles may be greater in drier years. A significant finding from this project was that crop growth rates are considerably lower than those expected for sngarcane. Measnrements of above ground biomass in the irrigation scheduling experiments showed rapid initial rates of crop growth until mid December (224 days after planting) for the plant crop and until early January (160 days after ratooning) for the first ratoon crop. However, after these samplings there was a substantial decline in crop growth rate. Comparisons with other areas show that early growth rates are not much lower than those measured in the Burdekin bnt later growth is significantly lower. The factors contributing to this reduction in crop growth rate from early in the wet season are not known. Muchow et af (1995) has suggested that stalk death associated with lodging can cause a slow down in yield accumulation in high yielding, long duration crops in Queensland, although the observed plateau in yield accumulation usualIy occurs at a higher cane yield and later in the growth period, some weeks after lodging. However in the Ord, the slow down in grOwt1l appears to occur independently of lodging and independently of crop start date, as shown by the results from the plant, first and second ratoon crops in block 2B. Further investigation of possible causes is clearly needed. Measurements of crop water use using EnviroSCAN equipment have shown that there is considerable variation during the growth of the crop. Water extraction by a sugarcane crop planted in May 1999 varied from a maximum of 10.9 mm/day during November to below 4mm/day from mid February onwards until harvest in June 2000. The evidence appears to suggest that there is a marked decline in crop water use to levels well below Class A pan 106 evaporation during and after the wet season. This occurs in both long duration, high yielding crops that have lodged and shorter duration, lower biomass crops that stay relatively upright. Clearly there appears to be a close relationship between the pattern of crop biomass accumulation and the rate of crop water use, with a marked slow-down in both occurring following the onset of the wet season. These findings have important implications for irrigation water management recommendations with more frequent irrigations being appropriate during the hottest time of the year in the October - December period and relatively few irrigations being needed following the wet season. This is particularly important as the main period when accessions to the water table appear to be occurring is during and immediately after the wet season. This post wet season period, when the soil profile is usually well supplied with water, is a critical period for making sure that irrigations are conducted only if absolutely necessary. Stalk elongation rates and pan evaporation losses are both used as indicators of when to irrigate sugarcane in parts of Queensland. These were tested as possible irrigation scheduling tools in the Ord but were found to support extremely frequent irrigations throughout the year which would most likely have resulted in the application of too much irrigation water. Clearly sugarcane crops in the Ord require the development of a special set of tools to assist the scheduling of irrigation. A key finding from this project has also been the high soil water extraction capability of sugarcane grown on Cununurra clay soils. Characterisation of four of the main soil types occurring in the Ivanhoe Plain has shown that plant extractable soil water (PESW) ranged from 167 mm on the heavy Aquitaine soils to 226 mm on Cununurra clays in the central Ivanhoe Plain. Given the much lower than expected crop water use post wet season, these very high values for PESW provide additional support for longer intervals between irrigations but maybe for more water to be applied at each irrigation. The four soils investigated are all different in both PESW and depth of water extraction and so recommendations for scheduling irrigation across the irrigation area will need to be fine tuned according to soil type. The information on soil water characteristics for each soil type is an essential prerequisite for the use of the APSIM sugarcane crop growth model for analysing the consequences of different irrigation regimes in order to develop strategies for best practice irrigation management. The APSIM-Sugarcane model has been used successfully in many environments to simulate crop biomass accumulation and crop growth rates. The preliminary work by Muchow and Keating (1998) to develop indicative estimates of irrigation requirements using the APSIMSugarcane model, using data available to them at the time, resulted in extremely high irrigation requirements approaching 40 MLiha/year, assuming an application efficiency of 60%. This figure was supported by the high annual water applications used by the Ord growers and documented in the 1995-96 survey of irrigation practices. However the APSIM model is not configured for any reduction in growth rate, such as that observed in Ord sugarcane crops. Consequently APSIM-Sugarcane overestimated the biomass of the plant and first ratoon crops in the irrigation scheduling experiment in block 2B by as much as 40% and would also have significantly overestimated their irrigation requirements. In order to rectify the problem the model was reconfigured, based on an analysis of the radiation use efficiency of these crops and the observations of Singh et al (2000) for crops which had lodged, to reduce the radiation use efficiency (RUE) to 1.0 g/MJ after the above ground biomass reached 30 tlha. With these modifications APSIM Sugarcane simulated less growth during the later stages of crop development and the simulations followed the observed growth patterns more closely in the field-grown plant and first ratoon crops. The modified simulations also reduced the crop irrigation requirement by about 25%. Whilst the modified 107 model describes the observed crop growth and biomass accumulation much more closely, further field investigation of the causes of the slow-down in crop growth is required before this phenomenon can be adequately represented in crop simulation models. Improving irrigation water application efficiency is dependant on minimising the amount of water losses associated with run-off from irrigation bays andlor losses due to deep drainage from irrigation bays or from infrastructure such as channels and drains. Opportunities for improving efficiencies have been examined in other irrigation areas by using surface irrigation models such as SIRMOD. Irrigation application efficiencies were investigated for a number of furrow-irrigated bays of sugarcane on Cununurra clay soils in the Ord. Data collected from each bay were used with SIRMOD to simulate application efficiencies for each bay. Whilst the correlation between simulated and actual values for particular bays was quite good, it was found much more challenging to configure the model so that it simulated irrigation events in a range of irrigation bays in the Ord. A further challenge was to acconnt for water infiltration rates associated with different furrow shapes, which can vary from a broad 'U' shape to a narrow 'V' shape. SIRMOD was eventually developed for general use in the Ord with both shapes of furrow. It was used to develop 'look up' tables indicating expected irrigation application efficiencies when such factors as moisture deficit, furrow inflow rate, irrigation duration, furrow length and furrow shape are varied. The look-up tables developed represent the most complete guidelines available for irrigators to use for maximising irrigation application efficiencies anywhere in the Australian sugar industry. A comprehensive set of best practice irrigation guidelines which maximise profitability and minimise losses of water are now available to the Ord sugar industry. These include: • Frequency of irrigation. This needs to be tailored to fit crop growth rate and crop water demand. Less frequent irrigations are required during early growth and more frequent irrigations during the hottest time of the year, particularly if the crop canopy is well developed. Very few irrigations are needed after the wet season due to the profile usually being filled with water during the wet season and because of the low crop water demand caused by the slow-down in growth. Soil water deficit should be used to schedule irrigations and irrigation frequency should also be varied for different soil types. • Soil moisture at irrigation is extremely important. Irrigating into dry soil minimises tail water flow and lengthening the period between irrigations is the best way to increase water application efficiency. Sugarcane is less sensitive to irrigation regime than was previously thought and is better able to recover during the wet season from moisture stress caused by insufficient irrigation prior to the wet than many other crops. • Tail water flow should not be allowed to run for too long and can be managed at different times of the year either by using different shut-off times or by varying siphon size. The duration of irrigation will depend on the time taken to wet up the soil around the cane plant. Often the soil surface never wets up but the soil below the surface can be quite wet. • Furrow shape will affect the time taken to wet up soils. An irrigation will advance down a bay much more quickly with 'V' shaped furrows than with 'U' shaped furrows. Consequently the latter will provide greater infiltration, less tail drain flow and better application efficiencies. However in the wet season, when good drainage is needed, 'V' shaped furrows are more effective in draining surface water off the bay. • Optimum row length. Longer row lengths will give better application efficiencies but poorer drainage and hence there is a trade off between the two. Shorter rows should be used on heavier soils to avoid waterlogging. 108 • Managing irrigations is much more difficult where deep drainage is occurring. Water losses in sandy areas require special management such as compacting on farm drains and channels. It is important to keep row lengths short in potentially leaky areas. Most deep drainage and accessions to the water table in the Ord appear to occur during and after the wet season when soil profiles are full of water. Despite all this knowledge on best practice irrigation management the number of irrigations and the amount of water applied by Ord sugarcane growers did not show much of a reduction between the benchmarking survey conducted in 1995-96 and the 2000-2001 survey. However there have been marked reductions in irrigation water use amongst some of the larger growers for the 2002 crop. This may be because the research team had not presented the growers with a clear message on irrigation management until the workshop held in 2001. Prior to that the growers had been exposed to large amounts of research results, some of which conflicted with established practices, but no clear management recommendations. It was not until the irrigation workshop in 2001 that the research findings were fully interpreted and translated into clear and unambiguous irrigation guidelines. Had the second survey been conducted on water management for the 2002 crop, the results may well have been somewhat different. Nevertheless there remains considerable scope for growers to make further improvements to their irrigation management practices and tailor their applications more closely to the requirements of their crops, and it wiII not be until this occurs that regional water tables in the Ord will start to stabilise. 109 15. REFERENCES Aldrick, J.M., Clarke, AJ., Moody, P.W., van Cuylenburg, M.H.R. and Wren, B.A. (1990). Soils of the Ivanhoe Plain, East Kimberley, Western Australia. Technical Bulletin 82, Department of Agriculture, Western Australia. Austin, N.R. and Pendergast, J.B. (1997) Use of kinematic wave theory to model irrigation on cracking soi!. Irrig. Sci., 18: 1-10. Banyard, R. (1983) ORIA Ivanhoe Plain Waterway Infiltration. Report No. WRB 57, Water Resources Branch, Kununurra, W.A. Baran, R., Bassereau, D. and Gillet, N. (1974) Measurement of available water and root development on an irrigated sugarcane crop in the Ivory Coast. Proc. Int. Soc. Sugar Cane Techno!., 15: 726-735. Bridge, B.J. and Muchow, R.C. (1982) Soil water relationships for Cununurra clays and Ord sandy loam in the Ord River Irrigation Area. CSIRO Aust. Div. Trop. Crops and Past., Trop. Agron. Tech. Mem. No. 30. Bureau of Sugar Experiment Stations (1982). Ord River Sugar Cane Project, Experimental Programme 1978-1982, Results and Recommendations. BSES Indooroopilly, Qld. Chan, K.Y. and Hodgson, A.S. (1981) Moisture regimes of a cracking clay soil under furrow irrigated cotton. Aust. J. Exp. Agric. Anim. Husb., 21: 538-542. Cox, W.J. and Chapman, A.L. (1985) Sugarcane. In Muchow, R.C. ed. Agro-research for the semi-arid tropics: North-West Australia. University of Queensland Press, St Lucia, Brisbane, 179-191. Dalgleish, N.P. and Foale, M.A. (1998) Soil matters. Monitoring soil water and nutrients in dryland farming. APSRU, Toowoomba, Qld., 122 pp. Gallagher J. (1986) Summary of sugar cane research 1950 - 1985. Marketing and Economics Branch, Western Australian Department of Agriculture. Gardiner, H.G. (1998) Sugar on the Ord: Getting it right. Australian Sugarcane, FebruaryMarch 1998,38-40. Gardiner, H.G. and Assoc. (2000) Ord Land and Water Management Plan 2000. Agriculture Western Australia, Bulletin 4446, 84pp. Graecen, E.L. (1981) Soil water assessment by the neutron method. CSIRO Division of Soils Report 76, 140pp. Gordon, L and Gardner, T. (1997) Prediction of deep drainage for irrigated soil profiles in the ORIA. Qld. DNR Report. Holden, J., Hussey, B., McDougall, A., Mallon, K. and Shannon, E. (1998) Water CheckImproving irrigation efficiencies in the Queensland sugar industry. Proc. Aust. Soc. Sugar Cane Techno!., 20: 126-132. 110 Inman-Bamber, N.G., Muchow, R.C., Holden, J.R., Robertson, MJ. and Ham, G.J. (1998) Soil water extraction by sugarcane beyond the readily available limit. Proc. Aust. Soc. Sugar Cane Techno!., 20: 112-117. Inman-Bamber, N.G., Robertson, M.l, Muchow, R.C., Wood, A.W., Pace, R.M. and Spillman, M.F. (1999) Boosting yields with limited irrigation water. Proc. Aust. Soc. Sugar Cane Techno!., 21 :203-211. Inman-Bamber, N.G., Zund, P.R. and Muchow, R.C. (2000) Water use efficiency and soil water availability for sugarcane. Proc. Aust. Soc. Sugar Cane Techno!., 22: 264-269. Keating, B.A., Robertson, M.J., Muchow, R.C. and Huth, N.I. (1999) Modelling sugarcane production systems. 1. Development and performance of the sugarcane module. Field Crops Research, 61: 253-271. Kingston, G., Hogarth, D.M., Albertson, T.O. and Freshwater, LT. (1980) Current research with sugarcane in the Ord. Proc. Aust. Soc. Sugar Cane Techno!., 2: 43-49. Kinhill Pty. Ltd. (2000) Water Infiltration Modelling for the Ord Sugar Project, Ord River Irrigation Area, Stage 2, Proposed Development of the M2 Area. Environmental Review and Management Programme, Draft Environmental Impact Statement. Kinhill Pty. Ltd., Victoria Park, WA. January 2000. Koehler, P.H., Moore, P.R., Jones, C.A., Cruz, A.D. and Maretzki, A. (1982) Response of drip-irrigated sugarcane to drought Stress. Agron. Journ>, 74: 906-911. Lee, L.C., Chapman, J.L and Basinski, J.J. (1963) Sugar cane trials at the Kimberley Research Station in North Western Australia. Aust. J. Exp. Agric. Anim. Husb., 3: 344-352. Leslie, J.K. and Byth, D.E. (2000) An analysis of sugar production issues in the Ord River Irrigation Area. SRDC Technical report No. 0112000. Lisson, S.N., Robertson, M.l, Keating, B.A. and Muchow, R.C. (2000) Modelling sugarcane production systems II: Analysis of system performance and methodology issues. Field Crops Research, 68: 31-48. McClymont, D. (1995) Infilt 3.01. User guide and technical documentation. University of Southern Queensland, Toowoomba, Qld. McGlinchey, M.G. and Inman-Bamber, N.G. (2002) Robust evaporation estimates of evapotranspiration for sugarcane. Proc. S. Afr. Sug. Techno!' Ass., 76: 245-249. Mitchell, A.R. and Van Genuchten, M. Th. (1993) Flood irrigation of cracked soi!. Soil Sci. Soc. of Am. J., 57: 490-497. Muchow, R.C., Evensen, C.L, Osgood, R.V. and Robertson, M.J. (1997a) Yield accumulation in irrigated sugarcane. II. Utilisation of intercepted radiation. Agronomy Journal, 89: 646652. Muchow, R.C. and Keating, B.A. (1998) Assessing irrigation requirements in the Ord Sugar Industry using a simulation modelling approach. Aust. J. Exp. Agric., 38: 345-354. 111 Muchow, RC., Robertson, M.l and Keating, B.A. (1997b) Limits to the Australian sugar industry: climatic and biological factors. In Keating, B.A. and Wilson, J.R. eds., Intensive Sugarcane Production: Meeting the Challenges beyond 2000. CAB International, Wallingford, U.K. pp. 37-54. Muchow, R.C., Spillman, M.F., Wood, A.W. and Thomas, M.R. (1994) Radiation interception and biomass accumulation in a sugarcane crop grown under irrigated tropical conditions. Aus!. J. Agric. Res., 45: 37-49. Muchow, R.C., Triglone, T., Plunkett, G.M., Sherrard, J.H., Wood, A.W. and Engelke, J.H. (2001) Field responses to irrigation in the Ord Sugar Industry. Proc. Anst. Soc. Sugar Cane Techno!., 23: 109-115. Muchow, RC., Wood, A.W. and Robertson, M.J. (1995) Does stalk death set the yield ceiling in high-yielding sugarcane crops? Proc. Aust. Soc. Sugar Cane Techno!., 17:142-148. Mnchow, R.C., Wood, A.W., Robertson, M.J. & Keating, B.A. (1996) Analysis of production potential and irrigation water requirements in the Ord sugar industry. Final report, SRDC Project ORD4S. Muchow, R.C., Wood, A.W., Spillman, M.F., Robertson, M.l and Thomas, M.R. (1993) Field techniques to quantifY the yield-determining processes in sugarcane: 1. Methodology. Proc. Aus!. Soc. Sugar Cane Techno!., 15: 336-343. Muchow, RC. and Wood, 1.M. (1981) Patterns of infiltration with furrow irrigation and evaporation of kenaf (Hibiscus cannabinus) grown on Cunnunurra clay in the Ord Irrigation Area. Aus!. Journ. Exp. Agric. Anim. Husb., 21: 101-108.. Nulsen, R.A and Sherrard, J.H. (1999) Water table monitoring in the Ord River Valley. Final Report, SRDC project WAAOO 1. O'Boy, C.A., Tickel, S.J., Yesertener, C., Commander, D.P., Jolly, P. and Laws, A.T. (2001) Hydrogeology of the Ord River Irrigation Area. Water and Rivers Commission, Hydrogeological record series, Report HG7, 80pp. Probert, M.E., Dimes, J.P., Keating, B.A., Dalal, R.C. and Strong, W.M. (1998) APSIM's water and nitrogen modules and simulation of the dynamics of water and nitrogen in fallow systems. Agric. Systems, 56: 1-28. Raine, S.R. and Bakker, D. (1996) Increased furrow irrigation efficiency through better design and management of cane fields. Proc. Aus!. Soc. Sugar Cane Techno!., 18: 119-124. Ratcliff, L.F., Richie, J.T. and Cassell, D.K. (1983) Field measured soil water availability as related to laboratory-measured properties. Soil Sci. Soc. Amer. Journ., 60: 593-600. Robertson, M.J. and Donaldson, R.A. (1998) Changes in the components of cane and sucrose yield in response to drying off before harvest. Field Crops Res., 55: 201-208. Robertson, M.J., Inman-Bamber, N.G. and Muchow, R.C. (1997) Opportunities for improving the use of limited water by the sugarcane crop. In: Keating, B.A. and Wilson, J.R ed. Intensive Sugarcane Production: Meeting the Challenges Beyond 2000. CABI, Wallingford, U.K. 287-304. 112 Robertson, M.J., Muchow, R.C., Donaldson, R.A., Inman-Bamber, N.G. and Wood, A.W. (I999a) Estimating the risk associated with drying-off strategies for irrigated sugarcane before harvest. Aust. J. Agric. Res., 50: 65-77. Robertson, M.J., Muchow, RC. and Wood, A.W. (1999b) A physiological basis for response of sugarcane to drying-off before harvest. Proc. Aust. Soc. Sugar Cane Techno!., 21: 196-202. Robertson, MJ., Wood, A.W. and Muchow, RC. (1996) Growth of sugarcane under highinput conditions in tropical Australia. 1. Radiation use, biomass accumulation and partitioning. Field Crops Research, 48: 11-25. Rose, C.W., Dayananda, P.W., Nielsen, D.R. and Biggar, JM. (1979) Long term solute dynamics and hydrology in irrigated slowly permeable soils. Irrig. Sci. 1: 77-87. Shannon, E.L. and Raine, S.R. (1996) Improving the irrigation efficiency of Burdekin Canegrowers. Proc. Aust. Agron. Conf., 8:502-505. Sherrard, J.H., Plunkett, G.M., Triglone, T., Bakker, D.M. and Engelke, J.H. (2002) Best practice water management for the Ord sugar industry. W.A. Dept. Agric. Internal Report, Research Activity Document, 95KU19. Sherrard, J.H., Riley, LT. and McCosker, P.F. (1991) Water management research for field cropping in the ORIA: 1987-1990. Internal Report, Western Australia Department of Agriculture. Sherrard, J.H., Robinson, C.J. and Plunkett, G.M. (1997) Groundwater Management in the Ord Irrigation Area. Paper presented at ISSCT Irrigation Workshop, Townsville, September 1997. Singh, G., Chapman, S.C., Jackson, P.A. and Lawn, RJ. (2000) Lodging - A major constraint to high yield and CCS in the wet and dry tropics. Proc. Aust. Soc. Sugar Cane Techno!., 22:315-321. Stoneman, T.C. (1972) Packsaddle Plains soil survey. Technical Bulletin 55, Department of Agriculture, Western Australia. Thompson, G.D. (1976) Water use by sugarcane. South African Sugar Journal, 60: 593-600. Walker, W.R. (1993) SIRMOD. Surface irrigation simulation software. Biological and Irrigation Engineering Department, Utah State University, USA. Wood, A.W., Muchow, R.C., Sherrard, J., Triglone, T. and Vogelsang, H. (1998) Benchmarking irrigation practices in the Ord Sugar industry. Proc. Aust. Soc. Sugar Cane Techno!. 20: 133-139. 113 ANALYSIS OF OUTPUTS AND OUTCOMES COMPARED WITH OBJECTIVES All of the original objectives of the project have been identified. These were as follows: I. Irrigation practices in the Ord Irrigation Area were benchmarked at the commencement of the project by means of a comprehensive grower survey. Any changes in practices were identified in a second survey conducted towards the end of the project. Whilst an industry workshop to define research priorities was not held, meetings to review research results and to plan further research were held twice yearly during the project and these allowed interaction between the research team and grower/industry representatives. Further review of research priorities occurred at Industry Board meetings, at Management Committee meetings for the Industry Development Officer and at the review conducted by SRDC. 2. On-farm and on-station research was successfully conducted to quantify the impact of different irrigation practices on productivity, water use efficiency and groundwater management. This research included drying-off experiments conducted both on-farm and on-station, irrigation scheduling experiments conducted on-station, soil characterisation conducted both on-station and on-farm, application efficiency measurements conducted mainly on-farm and measurements of crop water use and stalk extension conducted mainly on-farm. Direct analyses of the profitability of different irrigation practices were not conducted mainly because best practice irrigation management has only recently been defined. Some interaction has occurred between the Industry Development Officer and the Agriculture W.A. economist, Mr Francis Bright. It should be possible to quantify the water and labour savings associated with improved irrigation practices, altllOugh the environmental benefits of reduced accessions to the water table will be more difficult to quantify unless the costs of dewatering are used. 3. Sequential yield accumulation data was collected for the drying off experiments and for the irrigation scheduling experiments. Data from two automatic weather stations were also collected together witl! appropriate data on soils and crop management operations. These were all essential input data for the APSIM Sugarcane crop simulation model. The model, when used to simulate crop growth in the Ord, overestimated crop biomass and crop water use. Preliminary work to reconfigure the model with a reduced radiation use efficiency value for sugarcane with a biomass of >30 tlha produced much better agreement between simulated and observed values for crop biomass. However furti!er refinement of the model to tailor it for Ord conditions cannot really be done until the causes of the slow-down in crop growth are better understood. 4. Field experiments, databases and the APSIM Sugarcane simulation model have been used to help develop appropriate irrigation schedules for crops harvested or planted at different times of the year. The SIRMOD surface irrigation model has also been used to examine opportunities for improving water application efficiency and to develop a series of 'look-up' tables indicating expected irrigation application efficiencies when various parameters such as bay length, irrigation duration, inflow rate and soil deficit are varied. 5. Best practice irrigation management strategies have been developed and promoted by the project research team at Research Review and Planning meetings and at the 2001 Grower Workshop. The Industry Development Officer is responsible for ongoing promotion of industry best practice at field days, Industry Board meetings and grower workshops. A Best Management Practice booklet for the Ord, for distribution 114 to growers, is in preparation and this will include details of irrigation best management practices. Additional objectives for the project that were raised at the SRDC project review were: 1. Appointment of an Industry Development Officer. This was achieved on 15th March 1999 when Mr Jim Engelke transferred to Kununurra. 2. Soil variation across the Ord irrigation Area was recognised. Soil surveys conducted in the area recognised four main soil types: Cununurra Clay alkaline phase, Cununurra Clay eroded phase, Aquitaine Clay and Packsaddle Sandy Loam (Aldrick et at, 1990; Stoneman, 1972). Four soils in the Ivanhoe Plain were characterised with respect to their plant available water content and the likely depth of water extraction by sugarcane. 3. The early work on irrigation water application efficiency identified considerable variation between cane blocks. Even though blocks were carefully selected for detailed study, it was still a challenge to configure the surface irrigation model SIRMOD so that it simulated irrigation events over a range of different blocks. 4. Possible nutritional limitations on productivity improvement were included with the investigations conducted by the Industry Development Officer. These included a survey of fertiliser use by growers and measurements of nutrient removal in harvested crops from different farms. 5. Growers and industry representatives were consulted regarding the timing of the project review and planning meetings. A strong preference was expressed for the meetings to focus on the presentation of research results and to be held before the season (MarchiApril) and in NovemberlDecember after the season was finished. 6. With less than 20 growers in the whole Ord sugar industry, the need to establish smaIl focus groups of growers to facilitate the adoption of improved irrigation practices was not seen as a key objective. However the management committee set up to review the activities of the Industry Development Officer, with 2 grower members, 1 miller member, 1 CSIRO researcher and 1 from Agriculture WA, fulfilled the role of a small participative focus group. 115 POTENTIAL BENEFITS AND LIKELY INDUSTRY IMPACT The fDlIDwing benefits are likely to. arise from this prDject and impact Dn the sugar industries in the Ord and in eastern Australia: • A better matching Df irrigatiDn applicatiDns to. crDp requirements leading to. reduced use Df irrigatiDn water and reduced accessiDns to. grDundwater • Improved likeliho.Dd Df meeting future water licensing requirements and targets • Reduced IDsses Df irrigatiDn water thrDugh deep drainage and tailwater runDff. This in turn results in a reductiDn in the amDunt Df sediment, nutrients and agrDchemicals leaving cane blDcks. Deep drainage IDsses can be reduced Dnce prDblem areas are identified using EM68 surveys and irrigatiDn management practices are mDdified accDrdingly. • LDnger viability Df laser levelled cane blDcks and reduced siltatiDn Df tailwater drains. • Improved prDfitability with less frequent irrigatiDns through reduced irrigatiDn CDStS (labDur and water). • Improved sDcial impact through mDre leisure time and reduced stress. This can also. be achieved by using SIRMOD to. o.ptimise the duratiDn Df irrigatiDn events by changing bay lengths Dr siphDn size thus reducing the necessity to. turn Dff irrigatiDn water in the middle Dfthe night. • Reduced pressure Dn irrigatiDn infrastructure. MDre water available fDr additiDnal land areas. • StabilisatiDn Df grDundwater tables. Reduced need (and CDSt) for dewatering and Dther groundwater management strategies. • Meeting targets of Ord Land and Water Management Plan such as improvements in water applicatiDn efficiency, stabilisation of groundwater levels and improved water quality by reducing tailwater runoff. This will have the additional benefit of reducing pressures from environmental lobby groups. • The results Df this work will have direct benefits for other crops grown in the Ord such as cucurbits, chickpeas, sorghum and cottDn. The look-up tables for improving water application efficiency will be applicable simply by altering the soil mDisture levels required. Similarly the data frDm characterising different sDils in the irrigatiDn area will be applicable to. irrigation management strategies for Dther cropping systems. • ApplicatiDn of knDwledge gained from this project will have positive flow-on benefits for Dther irrigated sugarcane areas such as the Burdekin. 116 RECOMMENDED FUTURE RESEARCH NEEDS The following research needs to be undertaken in order to build on the major findings of this project: I. A major research challenge is to find reasons for the dramatic slow-down in cane growth. Further growth analysis experiments need to be conducted, particularly in drier years and on lighter textured soils to investigate whether the slow-down in cane growth is related to poor drainage, waterlogging during the wet season and loss of nitrogen by denitrification. Specific sections of different crop stands need to be carefully tagged so that stalk numbers in the same stools can be monitored both before and after the wet season, to test the hypothesis of Muchow et at (I995) that stalk mortality coupled with lodging of the crop are major contributors to the plateauing in growth occurring in many high yielding sugarcane crops. 2. Crop growth also needs to be monitored in crops planted and ratooned at different times of the year so that a better appreciation can be obtained of the timing of the slow-down in growth in relation to crop size and so that appropriate irrigation schedules can be developed for different crop start times. 3. The impact of deliberately slowing down the early growth of crops through withholding irrigation water warrants investigation in order to find out the extent to which the crop can recover during the wet season and whether it can attain the same yield as a crop which had more rapid early growth. The degree of catch-up in yield during the wet season is likely to depend on the amount of rainfall received. Improved skills in forecasting the likely onset and severity of the wet season for the Kimberley region through the use of indicators such as the Southern Oscillation index and Sea Surface Temperatures would greatly assist with the management of irrigation schedules for the October to December period and could lead to substantial water savings and reduced accessions to the water table. 4. A major obstacle to improved profitability in the Ord is the lack of knowledge of how to improve CCS and juice purity, particularly early in the harvesting season. Reasons for the lack of CCS response to drying off need to be found. Also experimental work with chemical ripeners needs to be undertaken to find out whether improvements are possible in early CCS and juice purity. 5. The APSIM-Sugarcane model needs to be reconfigured for Ord growth conditions as the current method of reducing radiation use efficiency (RUE) after crop biomass exceeds 30 tlha has little scientific basis except in lodged crops and is largely a way of achieving a better agreement between actual and simulated values for crop biomass. 6. The irrigation scheduling guidelines have not been adequately tested under commercial conditions in a range of different growing seasons. Further validation is required before growers can confidently adopt them for on-farm use. Similarly the look-up tables, based on soil water deficit, that have been designed to improve water application efficiency require further commercial validation. 7. The use of instrumentation on-farm is recommended to indicate the rate at which water is advancing down the bays. This will provide growers with greater confidence as to when to turn the irrigation water off and limit tail drain flows. The costs and benefits of tail water recycling should also be investigated as this is becoming a standard practice in some irrigation areas where water supplies are limited. 8. Areas where deep drainage is likely to occur need to be identified so that appropriate irrigation guidelines can be developed for these areas. Techniques are currently being investigated for mapping areas with a risk of deep drainage using EM3 I and EM38 instruments and this work needs to continue. 117 Some of these research needs are being addressed in a new SRDC project (CSE007 "Implementation of irrigation practices for profitable resource efficient sugarcane production in the Ord") that commenced in July 2002 and builds on the findings of this project. This new project aims to deliver the following outputs: • Information for implementing best management practices for optimising productivity and mill performance • Guidelines (rules of thumb) for best practice irrigation management developed, implemented and evalnated using an action research approach • Benchmarks of irrigation practice for sugarcane prodnction in 2002 and 2005 for evaluation of uptake of changed practices • Better definition of water reqnirements of sugarcane growing in the Ord as an input to water allocation policy • Processes based on action learning for implementation and evaluation of best practice irrigation management • An economic database on Ord sugarcane production costs and drivers • Model analysis of the economic and environmental consequences of different irrigation management strategies • Better nnderstanding of the physiological basis of yield accumulation for different growth stages and soil types under Ord climatic conditions as input to model analysis • Explanation of atmospheric drivers of crop evaporation using Bowen ratio as input to model analysis • Quantification of soil water extraction by sugarcane for different soil types as input to model analysis. 118 EXTENSION ACTIVITIES The following extension activities have been conducted during the course of the project to discuss research objectives and to promote and discuss the data collected and the ways in which it can be interpreted. Ord Sugar Industry Research Reviews The Ord Sugar Industry has been closely involved in the project from its inception. Every time the project team meets in the Ord, at least one meeting to review research progress is scheduled, to which all growers are invited together with mill staff and other interested personnel. It has been usual for around 60-80% of growers to attend each of the meetings, which are held in the late afternoon. Throughout the project growers have been extremely interested in the results and have asked many probing questions. They have also made a significant contribution to the way in which the research has been conducted and also to the way the results have been interpreted. The following topics were covered at the meetings: November 1996: Improving nutrient management of sugarcane, recent advances in furrow irrigation and irrigation efficiencies, chemical and biological limits to sugarcane production. March 1997: Groundwater management issues, findings from the grower survey of irrigation and crop management practices, update on crop logs, presentation on variety work and methodologies. October 1997: History of sugarcane research in the Ord, variable pol in cane, unique climate of the Ord, research progress in irrigation management project. May 1998: Overview of 1997 season and comparisons with 1996, presentation on irrigation management and plant extractable soil water, sngarcane nutrition in the Ord and new cane variety and pathology project. August 1998: Smut outbreak, factors contributing to low pol, climatic conditions in 1998 and impacts of nitrogen on productivity and pol. November 1998: Indnstry performance, mill operational issues and logistics, research into the reasons for low pol in the Ord, drying off and irrigation schednling trials, proposed survey of grower fertilizing practices, and forthcoming grower benchmarking study. April 1999: What is ccs? - implications of variation in fibre, brix and pol and varietal trends, possible canses of low pol in the Ord, factors affecting sugar yield on the Ord River, and developing nutrient management plans. September 1999: Current research and extension activities, grower feedback on key issues, update on variety issues and progress on crop nutrition guidelines, update of results from the best practice irrigation project. March 2000: Report on current extension activities including gappiness survey in plant crops and grower feedback on key issues, update on variety introductions and preliminary results from smut trial, report on results from the best practice irrigation project and possible 119 interpretation of the crop growth and sugar accumulation data, and presentation on water use efficiency results and ways in which application efficiency can be improved. November 2000: Overview of SRDC funding, update on irrigation research, related water research in Queensland including shallow groundwater work and irrigation risk management and crop water nse in the Bnrdekin, Ord research to improve irrigation application efficiency, variety update, and smut and other research. April 2001: Industry irrigation workshop (see below). November 2001: Results from irrigation scheduling experiment in block 2B, update on progress with APSIMIgrower treatments in irrigation trial in block 9A, review of mill commercial data (1996-2001). Ord Sugar Industry Irrigation Workshop A grower workshop was held with sugar industry representatives in the first week of April 2001 to examine ways of implementing local and other research fmdings relating to irrigation. Some notes taken at the workshop are given below: Irrigation Best Practice for sugarcane production in the Ord Irrigation Area Present: WiIlem Bloecker, Andrew Kelly, Gordon Mock, George Gardiner, Fritz Bolten, Lindsay Innes, Peter Letchford, Ray Cummings, Greg Cummings, Rocky Lerch, Barry Lerch, Mike Moore, Jim Engelke, Joe Sherrard, Derk Bakker, Geoff Inman bamber, Gae Plunkett, Tim Triglone, Peter McCosker, Andrew Wood, Sarah Strutt (facilitator) Objectives: • To increase awareness of irrigation best practice • To gain acceptance of irrigation guidelines as being practical • To obtain a commitment by the growers to evaluate guidelines. Review of previous research (Jim Engelke) • Research over the last four years has shown no significant responses to different irrigation treatments or to drying off prior to harvest. • Estimates of crop water requirements from the APSIM crop growth model are 22.631.5 MLlha of applied water. • The survey of irrigation water use conducted in 1996 showed that growers used 15.6 - 53.8 ML/ha of applied water • APSIM crop growth model has been a useful tool for examining the consequences of different irrigation management strategies • Measurements of yield accumulation have shown a distinct slow down in crop growth. The APSIM model has been reconfigured to fit the measured data • Growth slow down means much lower crop water use for later growth period after wet season • Irrigation guidelines suggest more frequent irrigations for early period of crop growth and less frequent irrigations after wet season • Water use estimates based on pan evaporation together with pan factors could be used to schedule irrigations Key messages: - Match water applied to crop demand - Irrigation guidelines need to be tested in the field 120 Recap of curreut research guideliues (Joe Sherrard) - Developing and implementing best practice for managing: • Irrigation application efficiency • Rising groundwater • Crop water requirements - Tail water flow 3-4 hours minimum to fully wet up soil. Do not need to run tail water for too long. - Lengthen period between irrigations to improve application efficiency. - Water required for crop growth less than initially thought due to growth slow down - Each hectare receives around 24 ML of irrigation and rainfall. Crop is only using around 17 ML/ha. Some of discrepancy is deep drainage. Key messages: - Manage frequency of irrigation post wet - Manage duration oftailwater flows - Optimise deficit at irrigation - Identify where deep drainage is occurring and minimise it Discussion about research guidelines • Management of tail water - different shut off times at different times of the year unlikely to be adopted. May need to use different siphon sizes after wet season rather than having different shut off times • Different amounts of biomass in block will affect advance rates Key guidelines a) From APSIM tables • Watering less frequently after wet • Watering more frequently during dry summer • Watering more frequently for early cut crops prior to wet b) From lookup tables from SIRMOD to optimise application efficiency • Not letting tail water run for so long • Flatten furrow shape (bed shape is independent offurrow shape and should be formed to minimise dirt in cane) • Manipulating row length • Alter siphon size to suit requirements • Manipulating slope • Irrigating onto drier soil Comment: polyacrylamides assist subbing of water and could save a couple of irrigations c) Other guideliues • Use of soil moisture monitoring sites • Knowledge of deep drainage (permeability mapping) allows people to manage accordingly such as keeping row lengths short in leaky areas • Compacting on-farm channels and drains • Liaise with Agwest I CSIRO staff for irrigation timing tables • Manipulating crossfall to achieve more constant head in head channel Discussion of opportunities presented by research guidelines • Some easy to implement • Others require major changes eg slope, run length • There is no choice - Ord growers have to become more efficient users of water and will have to accept a reduced water allocation • This workshop presents a good opportunity to revisit what we are doing 121 • We are going to have limited water allocations • There will be rules that people will have to obey in order to limit groundwater accessions - licensing conditions and severe penalties • Expectation in the community that change wiIl occur • There are good financial reasons for becoming more efficient • Opportunity to identifY inefficiencies in current practices by using APSIM model or other computer models or look up tables • Likely to be licensing requirements to cut tail water by at least 50% - tail water recycling • Public money should be provided to reduce leakiness ofMI channel and Diversion dam • Community expectations must assist research funding • Pursuing adoption of more efficient practices wiIl help to lever more $'s for implementation • Less run off means less silt going into drains. Would keep water costs down. Siltation of drains could lead to penalties to cover cost of cleaning them out. Issues and problems that need to be addressed before implementing research guidelines: • Translating research into something simple that can be used easily. Accessible tools for growers. • Some farmers do not see change as a priority • Tail water flows very difficult to manage and predict. How can we do it? • Requirement by regulator to reduce tail water by 50%. Need to know on-farm where leaky areas are • Limited knowledge of deep drainage - very difficult to quantifY. Deep drainage is beyond the control of most growers. Need to factor this into water allocations unless it can be managed. • Deep drainage likely to come more from rainfall than from irrigation • To go to a new, higher level of management, we will need much better information. Use of SIRMOD, APSIM, evaporation measurement, soil type characterisation, measurement of deep drainage. "If you can't measure it then you can't manage it". • Information needed at management scale (not just from lone research experiment) in order to make better decisions • Information we have now will allow some things to be done now. Other things will have to be done later on. • Efficiencies need to be made in distribution system by Government instrumentalities. System will need to be 80% efficient - nowhere near this at present. • Need to look at plant's ability to use water and soil's ability to hold it (organic matter, soil structure) Workshop Groups I. How do we translate research information into tools that are useful and accessible to growers 2. Tail water management Actions from group 1 on tools I. Provide simple tables with date of crop start based on past average conditions. These will include average time intervals between irrigations assuming no rain (will need corrections for rain). Calendar with suggested frequency of irrigations (choice of severity of water stress). Action: Geoff by mid May 122 2. Provide training in use of computer models to look at different scenarios either through Jim or for home use. Use model during wet season to look at different scenarios using real data. Action: Jim to organise training session in mid November 3. System for recording operational data to feed into model. Action: Derk by mid May 4. Better understanding of different soil types and influence they have on irrigation schedules. Interpretation of existing data on soil types to pick up which soil provinces exist and what guidelines apply. Action: Joe and Jim collate and interpret what is cnrrently known by mid May 5. Make Sirmod look up tables available. Action: Derk and Joe 6. Produce a field guide for irrigated cane consisting of2laminated A4 sheets containing interpreted information from tables. Action: Working group couvened by Jim (Jim + grower + water person) by mid May Actions from group 2 on tail water 1. Eliminate algae from channels which blocks up siphons. Action: Ord Irrigation Commission 2. Implement field guide containing tail water recommendations. See action 6 above. 3. Investigate simple measuring devices to monitor start and finish times of tail water flows. Action: Gae 4. Investigate growth retardants to minimise lodging as this makes tail water monitoring very difficult. Action: Jim Where to from here? o Need to move towards Best Management Practice Guide for sugarcane. o Ord Co-operative is prepared to help with the costs of producing and publishing this. o Co-op can also help to fund the cost of providing loggers and better systems for monitoring tail water flow. o Launch of guidelines at Field Day in mid June. o Hold a mini workshop to assess progress on actions and to revisit other issues o Notes from this workshop to be circulated. Irrigatioll scheduling "Ute Guide" A double page handout for sugarcane growers in the Ord was produced after the Irrigation Workshop to provide concise, practical guidelines for growers on irrigation schedules that meet crop water demands on different soils at different times of the year. The handont is attached in Appendix 2. 123 ACKNOWLEDGEMENTS 1. We wish to thank the following organisations which contributed financially to the work in this report: • Sugar Research and Development Corporation • CSRLtd. • Agriculture Western Australia • CSIRO Sustainable Ecosystems (formerly CSIRO Tropical Agriculture) • Ord River Canegrowers • Ord Sugar Pty. Ltd. • Cheil Jedang Ord River Sugar 2. We wish to thank all the cane growers in the Ord River Irrigation Area for the important role they played in the project and in particular for their enthusiasm for the project, their participation at project meetings, their completion of survey questionnaires and for willingly making land available on their farms for experiments and assisting with irrigation arrangements. In particular we wish to thank Peter Pegg, Rob Boshammer, Greg Cummings, Rocky Lerch, Lindsay Innes, Wilhelm Bioecker, Gordon Mock and George Gardiner who were closely involved with the experimental work. 3. We wish to thank the staff at the Ord Sugar Mill for their participation at project meetings, their assistance with harvesting arrangements of experimental sites and in providing crop productivity and pol data from the mill database. We particularly wish to thank Russell Kirk, Ben O'Reilly, Wayne Paul, Neill Farmer and Col Smith. 4. We wish to thank the technical staff at the Frank Wise Institute for their assistance with the field and laboratory work, Naomi Mackee of CSR Sugar for helping to set up procedures for juice analysis and Mike Spillman for his assistance in setting up the automatic weather stations and with field sampling techniques. 5. Many people have provided useful comments on the work presented in this report and have participated in helpful discussions. We particularly wish to thank Peter McCosker, John Leslie, Brian Keating, Peter Thorburn and George Gardiner. 124 INDUSTRY DEVELOPMENT OFFICER Introduction and background With the inception of the Ord Sugar Industry in 1995, it was decided that support for growers would be required. To address this issue, funding was initially made available by SRDC for an experienced extension officer from Queensland to service the Ord Sugar Industry. With the completion of this project in 1997, SRDC, the Ord Sugar Industry and the West Australian Department of Agriculture agreed to provide funding to employ a full-time Industry Development Officer (Sugar Agronomist) to service the research, development and extension requirements of the Ord. The position was first advertised in early 1998, readvertised in late 1998 and eventually filled in early 1999 when Mr J.H. Engelke accepted the position. Mr Engelke commenced activities in the Ord in March 1999. Range ofactivities 1999-2002 The industry established a list of activities to be undertaken by the Industry Development Officer. These included the following: • Data management • Irrigation • On farm crop management • Managing composition of cane supply • Verification of research outcomes on farm • Collation of information and communication A key component of the position was management of the variety project. The discovery of smut disease in mid 1998 focussed national sugar industry interest on the Ord and the successful management of the disease. This issue occupied considerable time in managing and arranging necessary trials, reports and updates to the sugar industry. The smut screening work continues to operate in the Ord as an integral part of the variety introduction programme. Regular reporting to the Sugar Agronomist Management Committee documents the range of activities in which the agronomist was involved. Being a new industry, a wide range of issues arose from the production of cane through to the processing environment. The End of Season reports provide a comprehensive list of activities that were undertaken during the 1999-2002 period. Outputs and Industry outcomes Many extension events occurred over the 1999-2002 period. These ranged from simple AgMemo articles through to detailed journal papers, research reviews and field days. These included: • 1999 Workshop (xl) • 1999 Field days (x2) • 2000 workshop (x2) • 2000 field days (x2) • 2001 Workshops / Research Reviews (x2) • 2001 Field Days (x2) • Irrigation scheduling 'ute guide' developed. • Commercial performance report - Analysis of mill data 1996-2001. J.H. Engelke. 125 • Conference Paper: Field Responses to Irrigation in the Ord Sugar Industry, by R.C. Muchow, T. Triglone, G.M. Plunkett, J.H. Sherrard, A.W. Wood and J.H. Engelke. Proceedings of the Australian Society of Sugar Cane Technologists 2001. • Conference paper: Sugarcane smut: successful management in the Ord, by J.H. Engelke, B.T. Egan, J.H. Sherrard, T. Triglone and P.A. Jackson. Proceedings of the Australian Society of Sugar Cane Technologists 2001. • Poster presentation: Ord Sugar Industry Statistics, by J.H. Engelke. Proceedings of the Australian Society of Sugar Cane Technologists 200 I. • Journal paper: Improving irrigation for Ord sugarcane, by J.H. Engelke, J.H. Sherrard, G.M. Plunkett and T. Triglone. Journal of Agriculture 2001. • Posters: Ord Sugar Industry Marketing and Sugarcane Quarantine. General discussions and conclusions The Sugar Agronomist Management Committee (now renamed Research Development and Extension Committee) provided a strong base and clear direction for the agronomist position. The requests of the Committee and industry to a large extent were for investigative work to be carried out that provided an indication of how the industry was running. In many cases this did not necessarily involve classic research bnt .rather information finding and reporting. Additionally, other more obscure issues such as nutrient mass flows through the sugar mill were required. The close working relationship between the miller and growers enabled rapid progress to be made on many issues without the interruption of any obstructive politics. It is particularly valuable when the mill, CJ Ord Sugar and previously CSR were only too happy to provide all information databases to facilitate investigations into crop production and quality to be performed and reported back to the entire industry. Being a new industry, many issues arose that required input from an agronomic point of view. In addition, as the industry is small with relatively few people involved, the agronomist was able to provide an independent view on many issues that arose between the growers, miller and harvest / transport contractor. Reports End-of-Season reports for the 2000, 2001 and 2002 seasons together with an End-ofFinancial Year report for the period 1999-2000 provide an outline of the various issues dealt with by the Industry Development Officer since March 1999. These are presented in Appendix 3. Field Days Five grower field days were held in November 1999, September 2000, July 2001, October 2001 and July 2002. Notes handed out at four of these field days are given iu Appendix 4. 126 APPENDIXl PUBLICATIONS, REPORTS AND EXTENSION NOTES Published Papers, Conference Papers, Reports, Field Day Notes, Ag Memos, Research Reviews and Research Workshops Published Papers Attard, S.J., Inman-Bamber, N.G. and Engelke, J.H. (2003) Irrigation scheduling in sugarcane based on atmospheric evaporative demand. Proc. Aust. Soc. Sugar Cane Techno!. 25: on CD. Engelke, J.H., Sherrard, J.H., Plunkett, G.M and Triglone, T. (2001) Improving irrigation for Ord sugarcane. Journal of Agriculture 42:45-50. Gosnell, J.M. and Engelke, J.R. (2002) Development of a sugarcane industry in the Ord, Western Australia. Proc. S. Afr. Sug. Technol. Ass. 76: 110-119 Muchow, R.C., Horan, H.L. and Keating B.A. (2003) Developing irrigation guidelines for the Ord sugar industry using the APSIM sugarcane mode!. Aust. J. Exp. Agric. (In prep) Muchow, R.C. and Keating B.A. (1998). Assessing irrigation requirements in the Ord sugar industry using a simulation modelling approach. Aust. J. Exp. Agric. 38: 345-354 Muchow, R.C., Triglone, T., Plunkett, G.M., Sherrard, lH., Wood, A.W. and Engelke J.H. (2001) Field responses to irrigation in the Ord sugar industry. Proc. Aust. Soc. Sugar Cane Techno!. 23: 109-115 Plunkett, G.M. and Muchow, R.C. (2003) Water extraction by sugarcane on soils of the Ord Irrigation Area. Aust J. Exp. Agric. 43: 487495. Wood, A.W., Muchow, R.C., Sherrard, J.H., Triglone, T. and Vogelsang, H. (1998) Benchmarking irrigation practices in the Ord sugar industry. Proc. Aust. Soc. Sugar Cane Techno!. 20: 133-139 Conference and Workshop Papers Bakker, D.M., Plunkett, G.M. and Sherrard, J.H. (1999) Improving irrigation water application efficiency in the Ord River Irrigation Area using computer simulation modelling. Irrigation 99 Conference, Perth, April 1999. Sherrard, J.H., Robinson, C.J. and Plunkett, G.M. (1997) Groundwater management in the Ord Irrigation Area. Paper presented at ISSCT Sugarcane Irrigation Workshop, Townsville, September, 1997. Sherrard, J.H., Engelke, J.H., Plunkett, G.M. and Triglone, T. (2001) Improving sugarcane irrigation in the Ord. Horticulture Program Workshop, Bunbury, September 2001. 127 Reports Engelke J.H. (2001) Ord Sugar Industry, Report on Commercial Performance 1996-2001. Department of Agriculture, Kununurra. Report Sherrard, J.H., Plunkett, G.M., Triglone, T., Bakker, D.M. and Engelke, J.H. (2002) Best practice water management for the Ord Sugar Industry. W.A. Dept. Agric. Internal Report, Research Activity Document, 95KU19. Industry Board Reports: 1999,2000,2001,2002 End of Season Reports: 2000, 2001, 2002 End of Financial Year Reports: 1999-2000 Sugar Check: Sugarcane Production on the Ord. Draft Best Practice Manual (in prep) Field Day Notes Field Day Notes - September 2000 Field Day Notes - July 2001 Field Day Notes - October 200 I Field Day Notes - July 2002 Irrigation Scheduling 'Ute Guide' (i) Sugarcane irrjgation in the Ord - Guidelines; (ii) Irrigation scheduling for sugarcane: three harvest dates & two soil water deficits AgMemos Sngarcane smut update, February 2000 Biobusiness Farming School, February 2000 Smut update, May 2000 ASSCT 2000 review, May 2000 2000 sugarcane variety trial results, August 2000 Compost Field day, September 2000 Smut trial update, October 2000 Red Stripe / Top Rot in sugarcane, November 2000 Smut trial plant crop ratings, April 2001 Sugarcane plant stand quality (1999 & 2000), April 2001 Sugarcane bed profiles (2000), May 2001 ASSCT 2001 review, May 2001 Sugarcane variety trial results (plant cane - 2000 planted trial), September 2001 Irrigation trial in block 9A, August 2002 ASSCT Queensland visit, August 2002 128 APPENDIX 2 IRRIGATION SCHEDULING "UTE GUIDE" 129 Total potential PESW Plant extractable water after irrigation with 80% refill (A x 80%) 50% total potential PESW remaining at irrigation (A x 50%) Water available for use by the plant (B-C) I 169mm 135mm 85mm 50mm 2 190mm 152mm 95mm 57mm Full canopy Full canopy to harvest IRRIGATION SCHEDULING FOR SUGARCANE THREE HARVEST DATES &: TWO SOIL WATER DEFICITS. Irrigation intervals are presented in days for each month in the tables below. ASSUMPTIONS & EXPLANATORY NOTES • Refill of the soil profile is assumed to be 80% of the total potential plant extractable soil water after each irrigation. • Irrigation scheduled after either 50% (Table A) or 40% (Table B) of total potential PESW remains in the soil profile. • Assumes no rainfall, therefore need to modify intervals when rainfall occurs. • Schedules are based on an average ofthe three soil types presented in Table I. • Long term average climatic data has been used to establish irrigation schedules. DD = Dry Down prior to harvest The three months after harvest have no entries as no real data are available (200 I irrigation trial examining this period). TABLE A: 50% of plant extractable water remains. JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC 1 AUGUST 9 TABLE B: 40% of plant extractable water remains. HARVEST JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC DATE 1 AUGUST APPENDIX 3 INDUSTRY DEVELOPMENT OFFICER END OF SEASON REPORTS END OF FINANCIAL YEAR REPORT, 1999- 2000 IimEngeIke Sugar Agronomist, Kununurra Data nlallage111ent CSR Data bases Obtained copies of previous years mill databases and set up for long term analysis purposes. Irrigation & Stalk elongation Stalk elongation measurements were taken on 3 paddocks during late 1999. (Q96 plant Cummings, Q96 plant Bothkamp, Q99 1R FWI) Q96 Cummings also looked at the relationship of stalk elongation in relation to available soil water. Results presented to industry December 1999. Assisted G. Plunkett in identification and establishment ofDUL / DLL sites in ORIA> Nutrition • Nutrient movement through the mill. A summary of the nutrient quantities exiting the mill is attached. These figures have been established by chemical analysis of mill mud and ash exiting the mill. More work is required to better understand what nutrients are plant available and the variation throughout the year. Calcium response trial established on Oasis Farms. Gappiness Survey • Plant cane gappiness survey completed and data compiled and presented to industry December 1999. Dirt in cane (Bedprofile survey) Results of the bed profile survey on plant cane compiled and presented to growers December 1999. Verification ofresearcll onfarm Dual row I single row (Oasis) • Two counts of the number of tillers/stalks per metre were carried out on Q99 dual row and single row cane. Next count harvest 2000. Results presented to industry December 1999. 130 Cane Variety Issues Observations of Q96 in the vaI!ey has seen the ratoons not looking healthy. Typically there are a number of stools which are doing weI! and others which have died and other not performing weI!. Established that Q96 is recognised for its sensitivity to harvester damage. (review of literature, inspection of affected paddocks) Cummings Q96 sampling over time for pol and fibre accumulation.(early 2000). Letter written to South Africa requesting access to cane varieties. Planting Frank Wise paddock 9A. Smut Issues • Smut trial establishment 1" September 1999. • Smut survey completed for the ORIA. Smut detected at Packsaddle in Nco-310 and Q96. Approximately 460 hectares has been plough out during 1999. Search of cane carried out. Nutrition • Sulphur Calcium trial established, Lerch • Cane Nutrient removal. Data collected from irrigation trial cane FWl. • Deep soil nitrogen survey. • Ash Analysis results obtained. No evidence was established of any deep soil nitrogen in either nitrate (N03) or ammonium (NH4) forms. Testing was carried out over 10 sites throughout the Ord River Irrigation Area to a depth of 2 metres. Results were presented to the sugar industry September 1999. Collation ofInformation and Communication Workshops I. Research review Presentation, September 1999. 2. Sugar Field Day / Update, December 1999. 3. Sugar Mill By product processing Field Day I" September 1999. 4. Bio Ag soil seminar. Field days / demonstrations 1. Kununurra Agriculture Show, Sugar Industry Display. 2. March Research Review 3. Sugar Industry Poster for Horticulture Review. Media 1. Report for the Round the MiI! publication (x2). 2. Ag Memo articles (x5). (Topic: 2 x Smut updates, ASSCT 2000, Bio Business School Feb 2000, China visit). 3. Article for the Primary Focus (x2).(Topic: China, smut) 4. Muresk Messenger (xl) 131 5. Australian Sugarcane Magazine (xl) (Topic: Smut in the ORIA). 6. ABC Radio interview with Greg Hayes, Kununurra (Topic: Smut) 7. ABC Radio Interview with Naomi Morison, Country Hour, Perth (Topic: Sugar Industry) 8. ABC radio interview on Ord Sugar Industry (general). 9. Numerous visits to sugar growers in the ORIA. Information maintain broad overview Initiated sugar cane manual to review and summarise relevant literature and information for Ord cane industry. Attended ASSCT conference Townsville, April 1999. Industry Board Requirementfor information Provided information to Canegrowers meeting on several occasions. Otlter • Glyphosate Damaged cane Bloecker. Report completed. • Leslie and Blyth report completed and submitted to industry November 1999. • Visit to Chinese sugar industry (Jan 2000) • Prepared end of season report (November 1999). • Completed Log frame, Horticulture Program Agriculture W.A. • Support for compost trial on 318 throughout 1999. • Quarantine protocol pursued and completed. • R. Lethbridge clearing proposal • K. Skerman soil maps/type information for subdivision application. • Assisted in sorting issues involved with the planting of9A. • A & S Mason. Paw -paw and banana inter-row cover crop trial. • CSR Mill induction. • Darryl Deacon supply information for teaching purposes. 132 END OF SEASON REPORT 2000 Jim Engelke Sugar Agronomist, Kununurra The following report is a summary of what occurred in Extension, Research and Development over the 2000 season. The report is separated into the three headings Extension Research and Development. The list relates to major activities ofthe Sugar R,D, & E through out the year. Extension: Collation ofInformation and Communication Information maintain broad overview Initiated sugar cane manual to review and summarise relevant literature and information for Ord cane industry. Attended ASSCT conference Bundaberg, May 2000. Industry Board Requirementfor information Provided information to Canegrowers meeting on several occasions. Workshops 1. Farming For Profit, February 2. Bio Ag soil seminar. 3. Nutritech soils seminar Field days I demonstrations 1. Kununurra Agriculture Show, Sugar Industry Display. (Sugar Industry poster & Compost poster). 2. March Research Review 3. Sugar Industry Poster for Horticulture Review. 4 Compost field day 31,1 August. 5 Smut trial field day 14°' September 6 Weed control field day (lAMA) 16'h November 7 November Research review. Media 1. Ag Memo articles (x8). (Topic: 4 x Smut updates, ASSCT 2000, China visit, Variety Trial results, Red stripe/Top rot). 2. Article for the Primary Focus (xl).(Topic: China) 3. Muresk Messenger (xl) 4. Sugarcane ORIA (Nitrogen, Cane Quality Measurement, Variety trial results, South African Link, Smut booklet). 5. Australian Sugarcane Magazine (xl) (Topic: Smut in the ORIA). 6. ABC Radio Interview with Naomi Morison, Country Hour, Perth (Topic: Sugar Industry) 7. ABC radio interview on Ord Sugar Industry (general) August. 8. Zimbabwean Farmer tours (x2). 9. Mozambique sugar industry representatives tour. 10. Qld politicians tour of mill / ORIA valley. II. Jerry Gosnell, South African agronomist tour of valley. 12. Numerous visits to sugar growers in the ORIA. 133 Research: Data management • Discussions with Horticulture Program manager (T.Hill) regarding funding availability for purchase of new wholestick planter and fertiliser rig. Two items purchased. • Enquiries regarding the purchase of a weigh bin. Funding issues to be resolved. Irrigation • Assisted G. Plunkett in the site selection and grower approval for establishing Drained Upper Limits / Crop Lower Limits in various soil types in the aRIA. Four DUL and three CLL were established for the aRIA. (One on FWI not included). • First ratoon crop successfully harvested and processed. Cane Variety Issues • Varity trial (R.V.T 1) harvested and results analysed. Three canes distributed for further 'on farm evaluation'. • Variety trial (R.V.T. 2) planted using smut trial results. Establislunent successful. • Introduction of39 new canes into open quarantine (levee block). • The new quarantine proposal for cane introduction reducing glasshouse time from 2 years to one year approved and finalised. • Currently in the process of seeking approval to place canes destined for the aRIA into the glasshouse facilities in Townsville rather than Brisbane in order to reduce costs. • Letter written to South Africa requesting access to cane varieties. Response received. Decision to be made on import possibilities. • Sent leaf material of 'Q125' to Queensland for verification of variety through DNA testing. No resuIts from testing to date. • Ripener trials on late cut Q99. No response detected. Smut Issues • Smut trial one completed. Plant and ratoon ratings compiled. • Smut trial two planted July 2000. Final plant rating to occur early 200 1. • Survey of selected commercial blocks of cane. Results support findings of smut trial one. • Australian Society of Sugarcane Technologists (A.S.S.C.T) paper on smut issues/progress in the aRIA currently being finalised. • Smut whip collection for BSES analysis pnrposes. Analysis performed in Canada for strain differences. Results not yet available. Development: Verification ofresearch on farm CSR Data bases • Obtained copies of previous years mill databases and set up for long term analysis purposes. 2000 database yet to be obtained. Cane Agronomy Issues • Cummings Q96 sampliug over time for pol and fibre accumulation.(early 2000). 134 o Hot water treatment plant operates for the first time. Treated a single batch of cane. o Monitoring the performance of three canes distributed for further evaluation MQ88-2047, Q125 and Q155. Irrigation & Stalk elongation o Stalk elongation work was pursued (G. Plunkett) on 3 blocks in the valley in conjunction with soil water information obtained using a 'Diviner'. Work is showing that stalk elongation can be maintained through more frequent irrigation, however, cane appears to be able to compensate if not water so frequently. Nutrition o Nutrient movement through the mill. Estimations of masses of various nutrients exiting the mill in by products. o Sodium mass flow carried out on ash water retum system utilised at the mill. Concluded that it is feasible that elevated levels of sodium are a result of concentration. o Nutrient removal determined for 5 paddocks in the ORIA. Sample analysis to be completed. o Leaf analysis data collection initiated. Approx 10 paddocks assessed for leaf nutrient status. Aim is to collate information on crop nutrient levels to more fully understand cane growth in the ORIA. o Obtained current fertiliser practice from eight growers in the ORIA. Using information to establish fertiliser costs and range of practices. Gappiness Survey o Plant cane gappiness survey carried out on 2000 plant cane and to be data compiled and presented to industry. o Weed pressure has appeared to increase from 1999 season. If this year is any indication of future trends weeds and in particular grasses, are likely to become a significant issue. Dirt in cane (Bedprofile survey) o Results of the bed profile survey on plant cane to be compiled and presented to growers. Generally bed profile looking more suited to harvesters requirements, still opportunity for improvement. Dual row / single row (Oasis) o Tiller / stalk counts completed on the single and dual row Q99 planted on Oasis farms. Little difference in final stalk numbers at harvest. Harvest results inconclusive. Continue to monitor the crops performance into ratoons. Nutrition o Sulphur trial (Lerch) harvested. Results indicate no obvious advantage from additional sulphur. o Cane Nutrient removal. Five blocks of cane sampled and to be analysed for nutrient removal. Testing yet to be completed. o Ash Analysis performed and results obtained. o On farm potassium trial established (Cummings) on 5th ratoon cane. 135 • Assisted in examination of options for by mill by product disposal. (Composting vs incineration) Other • Visit to Chinese sugar industry (Jan 2000) • Completed Log frame, Horticulture Program Agriculture W.A. • Support for compost trial on 318 throughout 2000. • Quarantine protocol for introducing new cane varieties pursued and completed. • R. Lethbridge clearing proposal • K. Skerman soil maps/type information for subdivision application. • Assisted in sorting issues involved with the planting of9A. • A & S Mason. Paw -paw and banana inter-row cover crop trial. • CSR Mill induction. • Darryl Deacon supply information for teaching purposes. Externalpart time study Post graduate study in Graduate Diploma Business (Agriculture) will be completed by mid 2001.0ptions include 1. Cease external study or 2. Continue into studying a Masters of Business. Approval is sought from the Agronomist Management Committee to continue with a Masters ifthat decision is taken over the next 12 months. 136 END OF SEASON REPORT 2001 Jim Engelke Sugar Agronomist, Kununurra The following report is a summary of what occurred in Extension, Research and Development over the 200 I season. The report is separated into the three headings Extension Research and Development. The list relates to major activities of the Sugar R,D, & E through out the year. Extension: Collation ofbiformation and Communication Information maintain broad overview Initiated sugar cane manual to review and summarise relevant literature and information for Ord cane industry. Attended ASSCT conference Mackay, May 2001. Two papers presented from the Ord, Smut and Irrigation paper. Industry Board Requirementfor information Provided information to Canegrowers and Industry Board. Workshops 1. Irrigation workshop, April 2001. 2. Edited and compiled Kununurra Horticulture Review papers, August 200 I. 3. Beef conference, Meeting Market Specifications, November 2001. Field days / demonstrations I. Kununurra Agricnlture Show, Sugar Industry Display. (Sugar Industry poster & Compost poster). 2. General sugar field day, July 2001 3. 3 x Posters on sugar industry for Horticulture Program Review, September 2001. 4 General sugar field day, 9A, October 2001 5 November Research review. Media 1. Ag Memo articles (x2). (Topic: 2 x Smut updates, ASSCT 200 I, Variety Trial results). 2. 2 x papers published in the ASSCT 2001 proceedings. 3. 1 x paper published in Journal of Agriculture. 4. ABC Radio Interview with Greg Hayes, Kununurra (Topic: Sugar Industry) 5. Jerry Gosnell, South African agronomist tour of valley. 6. Numerous visits to sugar growers in the ORIA. Research: Data management • Weigh bin received May and used in 2001 season. Irrigation • Purchased and installed a second Enviroscan (cane water use) • Second ratoon crop successfully harvested and processed. 137 Cane Variety Issues • Varity trial (RVT I) first ratoon harvested, results analysed and distributed. Three canes previously distributed for further 'on farm evaluation' bulked up on various farms. • Variety trial (RVT 2) plant crop harvested, results analysed and distributed. Establishment successful. • Modified cane introduction protocol to enable introduction of new canes direct from BSES glasshouses, Indooroopilly, Queensland. • Assisted in the access to overseas varieties held by BSES. • Introduction of 96 new canes into open quarantine (levee block). Smut Issues • Smut trial two completed. Plant and ratoon ratings compiled results distributed Nov 2001. • Smut trial three planted May 200 I. Final plant rating to occur early 2002. • Survey of selected commercial blocks of cane. Results support findiugs of smut trial one. Commercial planting of2002 not recommended. Development: Verification ofresearch Oil farm Mill Data bases • Obtained copies of previous years mill databases and set up for long term analysis purposes. First draft of report completed Nov 2001. Calle Agrollomy Issues • Inspection of Q96 indicates smut disease pressure increasing. As a result recommendation that no commercial planting ofQ96 in 2002. However, recommended that clean short hot water treated seed plots established. • Monitoring the performance of three canes distributed for further evaluation MQ88-2047, Ql25 and Q155. • Examined reasons for yield depression in 200 I. • Weeds continue to cause significant problems in many field of cane in the Ord. Nutrition • Nutrient removal determined for 5 paddocks in the aRIA. Sample analysis to be completed. • Leaf analysis. Approx 30 paddocks sampled for leaf nutrient status. Aim is to collate information on crop nutrient levels to more fully understand cane growth in the aRIA. Results available early 2002. • Obtained current fertiliser practice from eight growers in the aRIA. Using information to establish fertiliser costs and range of practices. Dirt in cane (Bed profile survey) • New butt lifter trial abandoned due to harvest contractor removing it from the machine before trialing was possible. 138 Dual row / single row (Oasis) • Greater number of stalks counted in the dual row in the first ratoon crop. However no yield difference. Nutrition • Cane Nutrient removal. Five blocks of cane sampled and to be analysed for nutrient removal. Testing yet to be completed. • On farm potassium trial harvested (Cummings) on 5'" ratoon cane. No yield or pol difference detected between treatments. Other • Completed Log frame, Horticulture Program Agriculture W.A. • Support for compost trial on 318 throughout 2001. • Quarantine protocol for introducing new cane varieties pursued and completed. External part time study Post graduate study in Graduate Diploma Business (Agriculture) completed 2000. Upgraded to a Masters of Business Administration (MBA) through the University ofNew England. Study will be completed end of 2002. Application for Department of Agriculture assistance has been lodged with the Regional Director, Dr Paul Novelly. 139 END OF SEASON REPORT 2002 Jim Engelke Sugar Agronomist, Kununurra The following report is a summary of what occurred in Extension, Research and Development over the 2002 season. The report is separated into the three headings Extension Research and Development. The list relates to major activities of the Sugar R,D, & E through out the year. Extension: Communication, facilitation and implementation Information maintain broad overview • Initiated sugar cane manual to review and summarise relevant literature and information for Ord cane industry. • Attended ASSCT conference Cairns, April 2002. • Visited the offices of the BSES (Mareeba, Brandon, Mackay, Brisbane) Impressed with the bio-tech programme. • Attended SASTA conference in July / August 2003. Impressed with ripener work developments. Industry Board Requirementfor information Provided information to Canegrowers and Industry Board. At request of Industry Board compiled Sugar Check (Draft stage) Workshops • Facilitated KPIA board meeting. Field days I demonstrations • 2 x 15 minute field days (9A irrigation updates & variety). • Research Review, April 2002. • November Research review 2002. Media • Ag Memo articles (x3). (Topic: I x Smut updates, ASSCT 2002, Variety Trial results). • I x paper published South African Association of Sugar Technologists. • Numerous visits to sugar growers in the ORIA. Research: Data management Irrigation • Purchased and installed a third Enviroscan (cane water use) for on farm demonstration purposes. • First ratoon 9A crop successfully harvested and processed. Cane Variety Issues • GxE trial harvested and data send to Queensland (no report available to date) • Variety trial (R.V.T. 2) IR crop harvested, results analysed and distributed. • R.V.T 3 established (6 series) 140 • Received another batch of overseas canes from BSES glasshouses, Indooroopilly, Queensland. • Introduction of 96 new canes into open quarantine (levee block). • Short hot water treated some Q96 (Oasis farms / Frank Wise) for use as plant material in 2003. Smut Issues • Smut trial 3 completed. Plant and ratoon ratings compiled results distributed Nov 2002. • Smut trial 4 planted August 2001. Final plant rating to occur early 2003. • Survey of selected commercial blocks of cane. Results support findings of smut trial 1, 2 and 3, with the notable exception ofMida. Recommendation for Mida remains that no commercial scale planting occurs. Development: Verification ofresearch on farm Mill data bases • Databases continue to be an important source of information for millers, growers and researchers. Mill management request large amounts of material accessible through the database system. • Compiled 2002 data into suitable format after W.Paul supplied the relevant databases. Yet to add 2002 data into the main database including all years. (Block paddock number changes make this difficult) Cane Agronomy Issues • Examination of root growth in plant cane has shown poorer that expected root system development. This was evident in both plough out replant and cane planted to virgin soil (Ex leucaena) • Wire worm damage detected in late planted cane (Mirrawong farm - Waugh & Dwyer) • Data collection on soil temperatures (IOmm and 50-75mm). Consitiuted part of the poor germination and root development investigations. • Inspection ofQ96 indicates smut disease pressure increasing. No commercial planting of Q96 unless plant material has been short hot water treated. • Varieties Mida, QI25 and Q 155 only constitute emergency canes, unlikely to be released for commercial production .. • Contributed to Ross Ridge report on 200 I season yield issues. • Weeds continue to cause significant problems in many field of cane in the Ord. Nutrition • Nutrient removal determined for 5 paddocks in the ORIA. Sample analysis to be completed. • Leaf analysis. Approx 30 paddocks sampled for leaf nutrient status. Aim is to collate information on crop nutrient levels to more fully understand cane growth in the ORIA. This continues from work started 200 I. Results available early 2003. Dual row / single row (Oasis) • Greater number of stalks counted in the dual row in the first ratoon crop. However no yield difference. 141 Other • Completed Log frame, Horticulture Program Agriculture W.A. • Support for compost trial on 318 throughout 2001. • On the by product disposal committee looking for options to dispose of excess mill bagasse, mud and ash. Study Lodged application for funding from the SRDC for funding to pursue PhD studies Comments Less was achieved in the position this year largely due to illness early in 2002. Leaving position in January / February 2003. 142 APPENDIX 4 INDUSTRY DEVELOPMENT OFFICER FIELD DAY NOTES 143 ORIAR&D [)RD SUGAR INDUSTRY FIELD DAY N[)TES SEPTEMBER 2DDD IRRIGATION Soil characterisation Crop water use Block 27 Irrigation trial results - Plant and Iratoon summary of results VARIETY UPDATE Replicated Variety Trial 1- Plant results Smut trial I- Plant HIratoon results Jim Engelke, Joe Sherrard, Gae Plunkett, Tim Triglone Department 01 Agriculture Sugar Research, Development tt Extension Kununurra WA 14/9/00 SUGAR CANE - CROP WATER USE ORIA R&D A series of sites around the valley have been selected to establish the total water holding capacity of the soil (Drained Upper Limit) and the amount of remaining after sugarcane has removed the available water (Crop lower Limit). The table below provides in millimetres the water present in 1 metre of soil (Drained Upper Limit) and the water remaining after extraction by sugarcane (Crop Lower Limit). The difference is the amount of water removed by sugarcane. Cununurra clay Aquitaine clay Drained Upper Limit (Total water, mm) 388 525 Crop lower Limit (mm) 268 381 Crop Extraction (DUl - Cll, mm) 120 144 Days taken for sugarcane to remove water. 129 68 Crop Water Use Plant Q96 Using a combination of a Diviner (electronic) and the Neutron Moisture Meter measurements estimates of a crop water use have been made. Crop Water Use Plant Q96 sugarcane, planted May 1999, harvest June 2000 12,-------------------------------------------------------. .g 10 ... Q) Co 8 Appoximate .............................................................................ti.at".Clt.l"d.9il1g............................................. J ----,;-----.---.------------------------_.--_.---------- -.------------.- ,6 .................................................................................., ........... ,................................................... j :.•••••••••••••.••••.••• ;~:;,:~~:,~:;•~,,:,:~',:":~.:":-~ E 0 m mm m mm mm 0 0 0 0 0 - - - - - - - - - -;;:: ....25m ~ m m m m m m 0 0 0 0 0 to CD co m ~N ~N to '"!2 !2 !2 - - - - -CD CD CD !2 CD 0 (0 ~ CD ~ ~ CD CD 0 CD !2 0 CD CD 0 (0 !2 CD ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ Date 14/9/00 ORIAR&D Assumptions and points of interest Assume the change in soil moisture at 20cm is the same as for 0-20cm. The evaporation rate over the period May 99 - Jun 00 was approximately 20% below the long term average. Total crop water use has been estimated to be in the vicinity of 17 megalitres The estimate of total water applied via irrigation to this crop was 14.5 megalitres per hectare (13 irrigations). Of the 14.5 megalitres applied it is estimated that 7-10 megalitres of this was used by the crop. VARIETY TRIAL RESULTS (MAY 2000) Plant Cane The 1999 variety trial was planted prior to information being available on the varieties level of smut resistance 1 susceptibility. As a result there are varieties in the trial that will be of little no commercial use based on smut susceptibility. BRIX = Total soluble solids in juice CCS = Commercial Cane Sugar TCH = Tonnes Cane per Hectare TSH = Tonnes Sugar per Hectare VARIETY BRIX CCS FIBRE% TCH MQ88-2047 18.8 13.2 14.9 187.5 PELORUS 18.8 12.8 15.8 189.2 Q127 18.5 12.7 15.6 190.4 Q96 20.2 12.4 19.9 187 KQ91-20301 19.6 13.6 18.9 165.8 KQ88-8075 19.2 13.7 15.2 162.7 Q99 18.1 11.6 14.6 192.2 MQ74-110 16.6 11.8 13.5 186.7 Q155 16.7 12.6 14.7 171.1 Q125 17.1 11.9 14.2 172.6 ORPHEUS 19.1 12.4 14.6 163.9 BMQ89-15 17 10.2 15.3 193.2 Q138 16.4 10.9 16.4 179.1 KQ91-31405 16.8 10.6 15.2 177.8 BMQ89-77 18.3 12.3 18.8 152.2 Q142 16.9 9.8 14.9 188.1 Q122 19.4 12.0 18.1 153.5 89-518-6 15.8 10.0 14.0 180.8 MQ97-155 14.5 9.0 14.5 200.8 89-503-10 16.6 10.1 14.2 175.7 KQ91-31506 16.6 9.9 13.9 178 Q161 14.6 9.6 13.5 181.4 89-680-3 15.6 9.8 16.3 174.5 BMQ89-14 14.4 8.7 13.7 184.8 MQ79-141 16.3 9.4 15.9 163.7 Q124 15.2 8.7 12.4 165 ** indicates varieties selected for further evaluation "on fann". TSH 24.66 24.22 24.11 23.20 22.59 22.32 22.28 21.98 21.57 20.54 20.25 19.79 19.57 18.82 18.75 18.47 18.38 18.10 18.01 17.71 17.58 17.48 17.03 16.17 15.43 14.28 ** ** ** 14/9/00 SMUT TRIAL RESULTS Latest inspection carried out 6th September 2000 ORIAR&D Results include the last rating of the plant crop (3/4/00) and the latest for the first ratoon crop (6/9/00) VARIETY 89-247-5 89-393-1 89-393-3 89-503-10 89-518-6 89-680-3 89-680-6 H73-6110 MQ80-805 MQ88-2022 MQ88-2047 Q125 Q130 Q155 Q95 Q99 Q142 90-83-5 KQ91-31405 ORPHEUS KQ91-31508 CP74-2005 Q145 90-77-5 H51-8194 84-608-10 Q135 PELORUS Plant 3104/00 4.2% 0.0% 0.0% 41.9% 0.0% 2.8% 0.0% 0.0% 0.0% 0.0% 8.3% 0.0% 0.0% 0.0% 1.6% 0.0% 0.0% 2.8% 48.4% 14.6% 41.7% 12.9% 21.2% 16.9% 0.0% 19.4% 13.1% 44.2% 1R 6/09/00 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 3.1% 3.6% 7.8% 9.4% 10.6% 11.1% 14.1% 15.7% 16.7% 18.8% 19.4% 20.5% VARIETY Q96 Q124 Q101 Q120 Q161 H78-7234 Q107 MQ79-141 Q150 90-110-9 MQ87-540 MQ87-155 ROC-1 KQ88-8075 Q137 84-608-3 88-402-2 Q113 87-105-10 84-608-6 BMQ89-77 Q162 Q138 BMQ89-14 BMQ89-15 89-605-1 KQ87-7339 NC0310 Plant 3/04/00 31.8% 41.7% 10.4% 40.0% 30.9% 62.2% 18.5% 57.0% 90.8% 41.7% 27.8% 38.2% 61.7% 75.4% 78.1% 34.7% 70.8% 40.3% 43.3% 75.0% 100.0% 85.4% 100.0% 78.1% 78.3% 33.0% 100.0% 58.8% 1R 6/09/00 24.9% 30.4% 34.4% 37.5% 44.0% 52.3% 52.5% 57.6% 58.3% 61.3% 63.6% 63.9% 66.7% 70.2% 71.9% 72.2% 72.2% 72.2% 72.9% 75.0% 75.0% 75.0% 75.6% 76.0% 77.0% 81.4% 84.4% 86.3% VARIETY Q164 MQ63-693 Q121 Q154 Q115 TS68-830 Tel/us MQ84-524 84-255-4 MQ74-110 Q117 90-77-2 MQ87-1069 84-255-10 86-1151-3 87-628-3 88-271-6 89-338-1 89-503-6 BMQ89-338 EOS KQ91-1003 Q122 Q127 Q157 Q158 Q159 Q91 Plant 3/04/00 61.5% 40.1% 71.4% 96.4% 64.0% 60.7% 97.2% 97.2% 100.0% 100.0% 91.3% 56.3% 100.0% 96.9% 94.4% 49.6% 97.5% 100.0% 86.1% 56.9% 100.0% 92.3% 100.0% 100.0% 100.0% 100.0% 100.0% 96.4% 1R 6/09/00 86.9% 88.1% 88.9% 90.0% 90.9% 91.0% 91.7% 91.9% 93.1% 93.8% 93.8% 94.7% 97.2% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% The results above are the second inspection of the first ratoon cane. The harvest date of the plant cane was the 1ih May 2000. Some ratings may vary from the plant cane final inspection. For example MQ88-2047 had an 8% infection level in the final plant cane inspection and in the latest first ratoon inspection above has 0%. No whip development has occurred. Variety H51-8194 has gone from 0% in the plant crop to 16.7% in the ratoon crop. This variety was severely affected by top rot in the plant crop. 14/9/00 IRRIGATION TRIAL Summary of results to date. PLANT CROP 099 (1998 -1999) Irrigations to canopy closure (5/5/98 - 18/9/98) Number Irrigations since canopy closure (1819/98 3/8/00) Amount of water applied (Megalitres) Cane Yield (tonnes per hectare) Pol (%) in cane 60mm evaporation 8 20 36 218.4 12.35 TREATMENTS 120mm evaporation 8 11 19.8 211.7 12.23 Sugar (tonnes per hectare) 26.97 25.89 180mm evaporation 8 7 12.6 198.4 12.98 25.75 ORIAR& D No significant difference No significant difference No significant difference FIRST RATOON 099 (1999 - 2000) Irrigations to canopy closure Number Irrigations since canopy closure Amount of water applied (Megalitres) 25-30 mm soil deficit 7 TREATMENTS 60-65 mm soil deficit 7 95-105mm soil deficit 7 16 6 2 27.3 10.5 3.6 "Soil deficit is calculated assuming there is approximately 300 mm of water in the soil to a depth of 80cm. 14/9/00 [JRD SUGAR INDUSTRY FIELD DAY NOTES JULY 20m IRRIGATION Water use Irrigation scheduling trials - current Hfuture New water use work Irrigation guidelines (Ute guide) Bed geometry and application efficiency work started. NUTRITION Nutrient removal and cane composition Tissue testing - 2000 results and 200! testing. Cane nutrition - Possible visit by Bernard Schroeder late 2001 VARIETY HSMUT SCREENING UPDATE Smut trial update - Smut trial 2plant inspection results OTHER W8eds in can8 Weigh bin BSES proposal Bug trapping Jim Engelke, JDe Sherrard, Gae Plunkett, Tim Triglone Department of Agriculture Sugar Research, Development !l ExtensiDn Kununurra WA IRRIGATION ~ Water use - EnviroSCAN and Diviner work. ~ Irrigation scheduling trials FWI. (Current & Future trials) Need 3rd treatment for future tria/. ~ New water use work - Need late cut, accessible, good stand of cane for annual water use study. ~ Irrigation guide and demonstration. ~ Bed geometry for water application efficiency work has started (no notes induded) Cane water use: The graph of water on B27 has been constructed using the "EnviroSCAN", which records changes in soil moisture content at pre determined intervals (15 minutes in this case). The soil moisture changes that result in water being removed from the soil are added for each day and a daily water use in millimetres can be established. Soil moisture can change as a result of; 1. Crop water use 2. Evaporation 3. Deep drainage For measurements at Block 27, the impact of deep drainage has been minimal and is less well defined for blocks 45 and 76. Water use figures are presented in table 1.0. WATER,USEBit'·· RI.~')tal1~'(1Rt996{§q~~(c~R~·./····· -2000-2001 (1R) Figure 1: Water use, as determined by change in soil moisture content, for a plant and first ratoon sugarcane crop (Q96) on Block 27 - 40. The next two graphs (Figures 2 & 3) of water use have been constructed using the "Diviner". Measurements of soil moisture are taken daily. Water use can be determined by summing the total of the changes in soil moisture. As a result of the single reading per day this method is not as sensitive as using the EnviroSCAN. Additionally the contribution of rainfall is more difficult to establish using the Diviner. Figure 2: Water use, as determined by change in soil moisture content, of a fifth ratoon sugarcane crop, Block 45. Figure 3: Water use, as determined by change in soil moisture content, of a first ratoon sugarcane crop, Block 76. NOT YET HARVESTED. Table 1: Crop water use estimated using soil moisture monitoringtechniques. Block Crop Water use Yield Water use (MUha) (tonnes cane Iha) (tonne cane/ML) 827 Plant 99/00 827 1R 00101 845 5R 00101 876* 1R 00101 * Block not harvested. 17.3 13.5 15.3 13.8 to 29/6/01 186 140 113 N/A 10.8 10.4 7.4 N/A Irrigation scheduling trials - update Current trial to be final harvest expected to be October 2001. Future trial beginning July 2001. 1. Using APSIM ( Irrigate when 50% of Plant Extractable Soil Water remains) 2. Using APSIM at ?? (drier treatment to be determined) 3. Grower established treatment. NUTRITION ; Nutrient removal and cane composition. ; Tissue testing - 2000 results, and 2001 testing. ; Tissue testing - Bernard Schroeder (BSES) visit possible towards end of 2001 Nutrient removal/usage A number of crops were sampled in the 2000 harvest season for nutrient content and removal / usage. Whole stalks were removed from the paddock and separated into stalk, cabbage, green leaves and dead leaves. Only dead leaves still attached to the stalk were recovered for weight and nutrient analysis. Composition ofsugarcane Using data collected from the nutrient removal work it is possible to generate information of the composition of sugarcane by component (stalk, cabbage, green leaves, dead leaves). Presented in the table below are the averages for the five crops analysed for nutrient removal! usage. STALK CABBAGE GREEN LEAVES DEAD LEAVES WET 86.7% 6.7% 5.1% 1.5% DRY 82.6% 6.8% 6.7% 3.9% Tissue analysis Several tissue samples were taken of crops late in 2000. These were analysed for nutrient content and compared to critical values established in Queensland and overseas. None of the samples taken displayed nutrient limitations using the critical values. Some of the samples taken are presented in the table below. It is planned to take more samples this year. NUTRIENT CRITICAL NUTRIENT VALUES" ;1,.' ,'" ,':,:,": ',",' 'I,:: :"': ':,L/'" •:1i1'MaC" "',' "',, "',",' " 1.8 0.19 1.1 0.13 0.18 0.08 UNIT Dry Weight % % % % % % P Q99 2.01 0.27 1.46 0.16 0.29 0.23 5R Q99 2.43 0:1"9 1.29 1R Q96 100kg/ha No Potassium Potassium 2.23 2.39 2.62 -O~24 0.23 0.24 1.51 1.52 1.57 0.15 0.16 0.16 0.16 0.25 0.36 0.37 0.38 0.18 0.24 0.18 0.18 2-3 mg/kg 5.6 4.7 6.1 5.8 6.1 10-12 -;"g/kg 15:6 18.5 -15.8 23.1 23.4 . . 'i~ib 15 0.08 mg/kg mg/kg 53 0.3 48 0.4 51 79.7 86.7 0.3 0.9 1.2 50 mg/kg 76 91 89 95 100 1-2 mg/kg 9.5 5.8 9.7 16.4 16.2 •• Plant nutrient concentrations need to maintained above the 'critical nutrient values' to ensure no nutritional limitations. The critical levels in this table are based on information from other cane growing areas. VARIETY & SMUT SCREENING UPDATE Smut Trials Smut Trial 1 - Planted September 1999: Completed plant and first ratoon rating Smut Trial 2 - Planted July 2000: Completed plant rating Smut Trial 3 - Planted May 2001: SMUT TRIAL 2 - Mean Infection at final plant crop inspection (30/3/01), sorted by level of infection. Variety 89-247-5 89-393-1 89-393-3 89-518-6 89-680-3 89-680-6 90-83-5 95H4004 95H4021 95H4024 95H4039 CP74-2005 K088-8151 M088-2022 M088-2047 ORPHEUS 0125 0130 0142 0155 0171 Mean Infection 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% Variety z-099 95H4040 90-77-5 0101 95H4001 95H4035 0107 0145 95H4005 95H4047 95H4007 0135 95H4029 95H4048 0161 z-096 95H4022 95H4010 84-608-10 95H4003 95H4033 Mean Infection 0.0% 3.3% 8.3% 13.2% 15.0% 15.6% 22.0% 22.2% 24.5% 25.7% 26.7% 26.7% 27.0% 27.6% 31.5% 36.8% 38.0% 45.6% 48.6% 50.1% 63.3% 95H4030 95H4016 95H4044 95H4032 95H4023 95H4008 95H4027 67N3184 K091-2616 95H4018 z-NC0310 95H4006 95H4012 95H4017 95H4020 95H4037 95H4046 0141 0165 0179 z-0117 63.8% 66.7% 66.7% 69.8% 74.4% 75.8% 76.7% 79.2% 83.8% 88.9% 95.2% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% z-095 0.0% The results of smut trial 2 plant ratings confirm those of smut trial 1. The final rating of the ratoon crop will be carried out later in 2001 (Nov). OTHER -+ Weed pressure in sugarcane -+ Weigh bin purchased and operational. -+ BSES proposal being pursued. -+ Bug trapping being carried out in ORIA sugarcane crops 2001. Weeds - Residual control In 2000 the weed pressures were noticed by a number of people to have increased in the sugarcane fields. Both broadleaf and grass weeds were observed to have increased generally across the ORIA. Some options for residual control of broadleaf and grass weeds on early emergent cane include: TREATMENT RATE/ ha Atrazine + Ametryn (Gesapax Combi) Atrazine + Diuron Atrazine + Flame 3L + 3L ( 6-8L) 2.2kg + 1.2kg 2.2kg + 400ml Atrazine + Stomp 2.2kg + 3L Atrazin + Trifluralin CR 2.2kg + 4.5L Atrazine + Balance 1.0kg + 200g APPROX COST -$60/ha COMMENTS -$48/ha -Flame price No information on performance in not available in aRIA. Avoid on sandy soils. aRIA -$63/ha -$68/ha Limited information on performance in the ORIA. -$64/ha Not for use on plant cane. "Check rates and prices and suitability with chemical sales organisation. []RD SUGAR INDUSTRY FIELD DAY NOTES []CT[]BER 2Dm IRRIGATION SA New irrigation trial Irrigation guidelines (Ute guide) VARIETY UPDATE Replicated Variety Triall-I Ratoon results Replicated Variety Trial 2- Plant results 2001 cane introductions OTHER Solar radiation investigation Jim Engelke, Joe Sherrard, Gae Plunkett, Tim Triglone Department of Agriculture Sugar Research, Development Ii Extension Kununurra WA I Map SUGAR CANE -IRRIGATION TRIAL 9A Frank Wise Research Institute """"'""" ""'"""" ""'""'5~ 1:: .-'5 '"" "15 0 ~" ", 15 ~'" il ~ '" '" '"~~08 0 0- 6?- 0 1.-<:.: '.".. ~ ~",'P":l .8 .'".. .<;>;:' 6?- 0 ~ ~ ~ "0 ~ 0 1-<:: Se'" ~~ "~[':''E' 00-g '.".. .<;>;:' ~ 0 ~ ~ ~ ~ ~'" 15 il ~ 8 CJ •.0;j " 1:! :>l .;";:' ~ 0 ~ ~ ~ ~ -'.i="..I ~ ~ "0 ~ 0 -< Se ~~ "~[':''E' white red yellow red yellow white yellow white red 2 13 13 23 2 1 HEAD DITCH Crop Log PLANT CROP Planted Harvested Yield 20 June 2000 5 July 2001 149.8 tonnes per hectare 1 RATOON Herbicide Ist irrigation Fertilised 2"' irrigation 3" irrigation (days since last irrigation) 4th irrigation 5th irrigation Future irrigation schedules Grower Irrigation Treatments APSIM (50% PESW APSIM (10% yield remaining) reduction) 6 July 2001- Gesapax combi 8 July 2001 7 August 200 I 10 August 200 I 31 Aug 2001 (21) 5 Oct 2001 (55) - 23 Oct 01 (73) 18 Sept 2001 (18) 2 Oct 200 I (13) 12 day intervals up to December 7 day intervals up to December 10-13 day intervals from 20 October to December 2 SUGAR CANE - VARIETY TRIAL RESULTS (Plant cane - 2000 planted trial) Two variety trials were due for harvest in 200I. The plant cane was harvested in July 200 I. The first ratoon crop will be harvested in September 2001. Listed in dot below is a summary of the relevant crop husbandry. • Pre-fertilised: • Fertiliser mix: • Planted: • Irrigated: • Herbicide: • Fertilised: • Irrigation: • Harvested: Depth 150mm.each side of bed centre. DAP-90%, Zinc Sulphate (mono) -7%, Sulphur - 3% at 300kg/ha. 23"' and 24'h May 2000 On top of cane 25/5100 Watered @120mm. Evap. Atradex 900 @ 4kg/ha Urea @ 300kg/ha. 23/8/00....ridge up. Watered @ 70mm.evap 4'h and 5'h July 20001 The table below presents the plant cane harvest results from 2000 planted replicated variety trial, sorted by sugar yield per hectare. All canes included in this trial have been previously assessed for smut resistance/susceptibility. Both resistant (0-10%) and intermediate (10-20%) canes are included in the variety trials. For varying reasons several canes in excess of 20% smut infection have also been included in the trial. Current commercial varieties Q96, Q99 and Q10I continue to perform well. Mida (MQ88-2047) which is undergoing field evaluation in the ORIA also performed well. Variety 096 MIDA (M088-2047) 0101 099 M080-805 0135 K088-8151 0142 0130 84-608-10 89-393-1 0155 M088-2022 ORPHEUS 0171 0124 095 K091-2616 89-518-6 90-77-5 0125 89-393-3 90-83-5 89-680-6 95H4035 89-247-5 Average LSD (P:1- - g~ (jJ t:! t:! '" 0~ CD Ol ""OJ" N ~ -CD Ol -3 CD '":l ~ 9 rows - no treatment Grower treatment D50 APSIM: Irrigate when 50% ofPESW remaining S10 APSIM 10% yield reduction D50 APSIM: Irrigate at 50% ofPESW remaining S10 APSIM 10% yield reduction Grower treatment S10 APSIM 10% yield reduction Grower treatment N D50 APSIM: Irrigate at 50% ~ ofPESW remaining RESEARCH STATION ROAD s: {l :r::r: :r: (f) " ":Im -'"OJ a~0 :ImD: 0 N Cil I ~ ~ III 3 "Q. m Ol :al. m 0 III CD Ol 3 "2. 5' co '" '" '$. () () (f) ~ aCD: a-m<:' -ma<:' aCD: Ol :al. oog~ 0 CJl ~ ""N :...., ~ W ~ ~ ".. C!J (jJ Ul ~C '" c;; '" '""" Gl ~ ~ a,~~ w IV :--' ~ 0 !" 01 N (f) '" '" '"'"p~ ~ ~ N ~ 0 CX> ~ 0 '" ~ '" '" '":0...., <0 0 (f) wm