Table Of ContentsNext Page

Improving the confidence of dryland maize growers in the northern region by developing low-risk production strategies

Michael Robertson1, Shayne Cawthray2, Colin Birch3, Rod Bidstrup4 and Graeme Hammer5

1CSIRO Sustainable Ecosystems / APSRU 120 Meiers Rd, Indooroopilly, Qld 4068, www.csiro.au Email Michael.Robertson@csiro.au
2
CSIRO Sustainable Ecosystems / APSRU 203Tor St, Toowoomba, Qld 4350 Email , Shayne.cawthray@csiro.au
2
The University of Queensland, Gatton Email c.birch@mailbox.uq.edu.au
3
Pioneer Hi-Bred, Dalby Email rod.bidstrup@pioneer.com
4
QDPI, Toowoomba and The University of Queensland, St LuciaEmail hammerg@dpi.qld.gov.au

Abstract

Highly variable rainfall in the maize-growing areas of the north-eastern wheatbelt has resulted in recent years in a shift away from maize production towards crops such as sorghum and cotton. This paper reports upon a research program to address the development of low-risk production strategies in maize. Two on-farm experiments are taken as case studies, one examining the response to supplementary irrigation, the other evaluating the risks of maize production relative to the mainstay sorghum.

Key Words

drought, population, sorghum, soil water, simulation modelling

Introduction

Highly variable rainfall in the maize-growing areas of the north-eastern wheatbelt has resulted in recent years in a shift away from maize production towards crops such as sorghum and cotton. Grower confidence has also suffered due to fluctuating market prices. There is increasing demand for maize as a feed grain and for silage from the expansion of the feedlot beef industry, particularly on the Darling Downs. In addition, there is a demand from grain growers to evaluate a wider range of summer crop options following the recent downturn in cotton prices, general variability in prices for sorghum and sunflower, and the problems with Fusarium in irrigated cotton lands. This paper reports on the use of participatory on-farm experimentation, supplemented with simulation modelling, to develop appropriate risk-minimisation strategies for dryland maize production.

Method

Two experiments, run in collaboration with growers, focussed upon contrasting roles for maize in dryland farming systems.

Experiment 1 was located near Dalby on the eastern Darling Downs in Queensland. In this environment, maize is often grown under supplementary irrigation on soils with high plant available water capacity (PAWC) (>200 mm), consequently yields are relatively stable from season to season. The presence of other high value dryland crops that compete for the available irrigation water means that increases in yield and return from maize with irrigation have to be compared against alternative uses for the water. In this experiment, cv. 3237 was grown under three water regimes: dryland, one in-crop irrigation, and two in-crop irrigations. All regimes received an irrigation before sowing on 4th October. In addition, a range of population densities were used, along with skip-row and solid plant configurations, within each water regime to explore the interaction between plant density and water supply. In-crop rainfall was 307 mm, compared to a long-term average of 387 mm.

Experiment 2 was located at Meandarra on the western Darling Downs in Queensland. In this environment, low and variable summer rainfall imposes considerable risks for dryland crop production. In addition, sub-soil constraints in the form of salinity, sodicity and acidity can restrict the availability of stored soil water to crop roots. In this farming system, rapidly rising air temperatures in spring and early summer coupled with high evaporative demand necessitates early sowing of maize to avoid damaging stresses around the time of grain set. Maize is commonly sown earlier than sorghum due to its superior ability to establish under cool temperatures. In this environment, maize is a largely untested crop, so in 2001-02 an experiment was conducted to assess the impact of plant population density on maize productivity, and to compare the productivity of adjacent maize and sorghum crops. Maize cv. 3237 was grown under densities of 20, 30 and 40,000 plants/ha in a double skip row configuration with 1.0 m between rows and a 2.5 m skip spacing. In addition sorghum cv. Buster was sown as a comparison. Characterisation of the soil PAWC and measures of plant available water (PAW) at sowing showed that the crop had a total of 114 mm out of a capacity of 231 mm. In an adjacent field, sown 8 days later, maize cv. 3237 was compared with sorghum cv. Bonus. Comparable values for PAW for this field were 50 and 161 mm, respectively. In-crop rain between sowing on the 3rd September and harvest in late January was 204 mm. The long-term average at Meandarra for this period is 314 mm.

Simulation analysis with APSIM (1) was used to extrapolate the finding in each experiment using long-term climate data from each locality.

Results

Response to supplementary irrigation

Experimental results at Dalby showed no yield difference between skip-row and solid plant configurations, however there was a 2 t/ha response to one in-crop irrigation of 75 mm and a further 2 t/ha response to an additional irrigation of 75mm (Fig. 1).

Figure 1: Simulated and hand- and machine-harvested grain yields for cv. 3237 at Dalby under four water supply-plant configuration regimes. There was no hand-harvest in the 2 irrigation treatment.

Hand-harvested yields were higher than machine harvested due to in-field variation and harvesting losses. There was good agreement between simulated and hand-harvested yield. The substantial response to irrigation was due to high evaporative demand and no rainfall during grain-filling.

Growers involved in this experiment asked “This was an impressive result, yet how often am I likely to see such responses?” “If my water becomes more expensive will it pay me to apply one, two or more irrigations?” In order to answer these questions, long-term simulations using the historical climate record at Dalby were configured using the same crop management, cultivar and soil characteristics as in the experiment. A density of 5 plants m-2 was used and solid (1 m rows) configuration was compared under dryland, 1- or 2-in crop irrigations. Simulations showed (Fig. 2) that, on average, a 1.5 and 3.3 t/ha response would be expected, slightly less than that seen in the experiment. With current maize growing prices and costs this would equate to a net return of $260 and $578 /ha. Growers saw this as a healthy

return upon the cost of their water and competitive against alternatives such as cotton.

Figure 2: Simulated increase in grain yield and net return at Dalby in response to 75 or 150 mm in-crop irrigation over the past 44 year climate record. Net return calculations assumes a maize price of 200 $/t and a water cost of $50/ML. Mean values are the columns on the extreme right of the graphs.

Table 1: Details of maize and sorghum comparison in adjacent paddocks near Meandarra.

Sowing date

6 September 01

14 September 01

Starting PAW (mm)1

102

50

 

Maize

Sorghum

Maize

Sorghum

         

Cultivar

3237

Buster

3237

Bonus

Established plant density (m-2)

2.1

3.9

2.4

4.0

Grain yield (kg/ha)

       

Hand harvest

2877

3063

1983

2232

Machine harvest

2345

3020

1445

2295

Simulated

2801

3205

2013

1886

1Sampled 21 September 01

Risks of maize production versus sorghum

Experimental results showed maize yielding 0.2 t/ha less than sorghum for the 6th September sowing, which had a starting PAW of 102 mm. The sowing in the adjacent field on the 14th September yielded considerably less, despite receiving the same rainfall, because starting PAW was only 50 mm. In this sowing maize also yielded around 0.2 t/ha less than sorghum. There was good agreement between simulated and hand- and machine-harvested grain yields.

Growers involved in the experiment asked “How does maize stack up against sorghum in the long term, given that maize would usually be sown a month earlier than sorghum and that they would both normally be sown with 100 mm PAW rather than 50 mm?” In order to answer these questions, long-term simulations using the historical climate record at Meandarra were configured using the same crop management, cultivar and soil characteristics as in the 6th September sowing, except that sorghum was sown on 6th October. Simulations showed (Fig. 3) that yield variability was greater for maize – in 10% of years no yield would be harvested, due to water stress around flowering, while in sorghum yield always exceeded 1.5 t/ha. Sorghum never yielded more than 3 t/ha, while maize yielded more than 3 t/ha in about one year in three. On average, for this scenario, sorghum yield was 2468 kg/ha while maize yield was 1921 kg/ha. At these yield levels the growers were satisfied that maize was competitive with sorghum because, while yielding on average less than sorghum, the use of maize in the system afforded other benefits such as a longer time to fallow after harvest before the next crop and the opportunity to utilise sowing opportunities when they occur in the spring, before sorghum could be sown.

Figure 3: Probability of exceedance distributions for yield of maize and sorghum for the past 44 year climate record. Maize cv. 3237 at 2.5 plants m-2 was sown on 4th September and sorghum cv. Buster was sown at 4 plants m-2 on 4th October, both under double skip configuration. Starting soil conditions each year was 110 mm PAW.

Conclusions

The research approach outlined in the paper illustrates the utility of combining on-farm experiments with simulation to address issues of riskiness of production in dryland cropping systems. In the two examples used, the issues of response to limited irrigation, and production risk compared to an established crop, were addressed. Simulation, once tested on the experimental results, allowed the experimental results to be confidently placed in a longer-term perspective, while allowing exploration of alternative strategies.

References

(1) Keating, B. A., Carberry, P. S., Hammer, G. L., Probert, M. E., Robertson, M. J., Holzworth, D., Huth, N. I., Hargreaves, J. N. G. H., Meinke, H., Hochman, Z., McLean, G., Verburg, K., Snow, V., Dimes, J. P., Silburn, M., Wang, E., Brown, S., Bristow, K.L., Asseng , S., Chapman, S., McCown, R. L., Freebairn, D. M., Smith, C. J. (2002). The Agricultural Production Systems Simulator (APSIM): its history and current capability. European Journal of Agronomy (in press).

Top Of PageNext Page