Previous PageTable Of ContentsNext Page

Using Modelling and Agronomy to Improve Reliability of Maize as a Rainfed Crop in Northern New South Wales and South Western Queensland

Colin Birch1,3, Kirsten Stephen1, Greg McLean2 and Graeme Hammer1

1 The University of Queensland, Gatton Campus, Gatton, 4343.
2
Queensland Department of Primary Industries and Fisheries, Toowoomba, 4350
3.
Present address: The University of Tasmania, Cradle Coast Campus, Burnie, 7320, Tasmania

Abstract

Expected trends in climatic conditions, prices for maize and other grains and development of grain based ethanol mean it is timely to reassess maize as an option for areas not traditionally considered for maize and compare predicted performance to that in established areas. APSIM-maize was used with long term weather records to model maize production in the Western Darling Downs (Queensland) and North Western Slopes and Plains (NSW) and for established and other potential production areas for combinations of 100 and 67 % plant available water (PAWC) at planting in representative soils, plant populations of 2 and 4 m-2, quick and medium maturity cultivar types, and planting on 15th August, 15th November and 15th February. There were wide ranges in predicted yields, from total crop failure to > 8 t ha-1 (Moree, Goondiwindi) and > 9 t ha-1 (Quirindi, Gunnedah). The analysis indicated that the greater risk of failure because of greater variability in lower rainfall, higher temperature areas could be managed by early or late planting into soils with 100% of PAWC using quick maturing cultivars at2 m-2 of. Medium maturity cultivars and higher plant populations were indicated for higher rainfall, moderate temperature areas in the study. Implications for agronomic practices and strategies to optimize efficiency of plant performance are discussed in relation to environmental conditions.

Introduction

The high yield potential of maize is not always realised under commercial conditions for a number of crop adaptation, environmental and agronomic reasons. Experimental yields in Australia are usually higher in established maize production areas with moderate temperature (e.g. Eastern Darling Downs, Liverpool Plains) than in northern and inland areas, most of which are hotter and drier. Maize production in areas of <600 mm annual rainfall that have relatively high temperatures is considered less reliable than grain sorghum (Routley and Robertson 2003) and has been studied experimentally and using modelling (Robertson et al. 2003, Routley and Robertson 2003). The increasing demand for maize for human consumption, stockfeed and ethanol production means increased production from new and existing areas is needed. Target yields for rainfed maize in southern Queensland and northern NSW range from 3 to 7 t ha-1 (DPI 1998, DPI 2002 cited in Birch et al 2003, O’Keefe, 2006), though experimental and commercial yields vary widely. The recent development of cultivars with comparative relative maturity (CRM) ratings as low as 95-97days means reassessment of options for maize production is appropriate. This paper uses the APSIM modelling framework (Keating et al 2003) which has been widely validated for dryland and irrigated maize for example, in Australia, United States, Kenya and India (Robertson et al 2003, Lyon et al 2003, Micheni et al 2004, Dimes and Revanuru 2004). It has been applied by these and other authors (eg Birch et al. 2008) in analysis of yield variation in a range of environments. This paper concentrates on assessment of predicted yield for combinations of maturity type, planting time and water availability at planting in vertosols in lower rainfall, higher temperature areas of Northern New South Wales and Southern Queensland, with some sites outside these areas included for comparison.

Materials and Methods

APSIM - maize was used with long term weather records (around 100 years) to model yield of maize grown in current and potential maize production areas in Queensland and New South Wales. Plant available water capacity (PAWC) to 1.8 m in the APSIM soils data base for representative soils in each locality was used eg 150 mm (Goondiwindi), 190 mm (Moree) area and 290 mm at Gunnedah and Quirindi). Combinations of 2 and 4 plants m-2, quick and medium maturity cultivar types, planting dates of 15th August, 15th November and 15th February, with 100 and 67% of PAWC at planting were used.

Results

Table 1. Predicted median yield (PMY, t ha-1) for planting dates, cultivar types, plant populations and proportions of PAWC to 1.8m at Chinchilla (Ch), Roma (R), St George (StG), Gondiwindi (Gw) (SW Queensland), Moree (Mr), Gunnedah (Gd) and Quirindi (Qd) (NW New South Wales).

(a) quick maturity cultivar, 2 plants m-2, 100 and 67% of PAWC

PMY range (tha-1)

Planting 15 August

Planting 15 November

Planting 15 February

100%

67%

100%

67%

100%

67%

<2.0

StG

Rm, Gw

StG

Rm, Gw

Qd

Qd

2.0 – 3.0

Rm, StG

Mr

Gw, Rm, StG

Mr

Gd

Ch, Rm, Gw, Mr, Gd

3.0 - 4.0

Ch, Gw, Mr

Ch

Ch, Mr

Ch

Ch, Rm, Gw, Mr

 

4.0 – 5.0

     

Gd

StG

 

>5.0

Gd, Qd

Gd, Qd

Gd, Qd

Qd

   

(b) medium maturity maize cultivar, 2 plants m-2, 100 and 67% PAWC

PMY range (tha-1)

Planting 15 August

Planting 15 November

Planting 15 February

100%

67%

100%

67%

100%

67%

<2.0

Rm, StG, Gw, Mr

Ch, Gw, StG,
Mr

StG

Ch, Gw, StG,
Mr

Ch, Mr, Gd,
Qd (0)

Gw, Gd,
Qd (0 )

2.0 – 3.0

Ch

 

Ch, Rm, Gw,

Mr

 

Rm, StG, Gw

Gw

3.0 - 4.0

Gd, Qd

Gd, Qd

 

Gd

   

4.0 – 5.0

   

Gd, Qd

Qd

   

(c) medium maturity maize cultivar, 4 plants m-2, 100 and 67% PAWC

PMY range (tha-1)

Planting 15 August

Planting 15 November

Planting 15 February

100%

67%

100%

67%

100%

67%

<2.0

Rm, Gw, Mr

Rm(0), Ch(0), Gw, Mr

Rm, Gw, Mr

Rm(0), Gw(0), Mr

Rm, Gd, Qd (0)

Rm(0), Gd, Qd(0)

2.0 – 3.0

       

Mr, Gw

 

3.0 - 4.0

 

Gd, Qd

 

Gd

   

4.0 – 5.0

     

Qd

   

5.0 – 7.0

Gd, Qd

 

Gd, Qd

     

Early planting produced at least as high predicted yields as later options at most sites, though planting in February produced among the highest PMY at some sites (eg St, George, Moree).

Discussion

Predicted median yield (PMY) ranged from zero to >8 tha-1 for quick cultivar types at 4 plants m-2 planted on 15 th November in an established production area (Gunnedah and Quirindi, Table 1). Wide variation in predicted yield is shown by the greater range and lower slopes in the cumulative distribution functions (CDF) for maize planted at 2 plants m-2 in lower rainfall areas (Goondiwindi, Moree, Chinchilla, Roma, St George) than more favoured environments (Figure 1). However, late planting in cooler areas (eg. Quirindi) changes the shape of the CDF and reduces PMY. Planting on 67% PAWC eg 50-60 mm less PAWC in Goondiwindi, Moree, Roma and St George areas. reduced predicted yields at most sites and increased probability of crop failure (yield <2 t ha-1 , Birch et al 2008a). Higher plant population did not increase PMY except in favoured areas eg Quirindi but increased probability of failure (Table 1).

Figure 1 Cumulative distribution functions for predicted yield of a quick maturity type grown under differing conditions at several locations in Queensland and northern New South Wales

(a) planting Date-15th August; Population - 2 m-2; Plant Available Water – (L) 100% (R) 67%;

(b) Planting Date:15th November; Population 2 m-2; Plant Available Water – (L) 100% (R) 67%;

(c) Planting Date:15th February; Population 2 m-2; Plant Available Water – (L) 100% (R) 67%;

The analysis indicated that quick maturity cultivars are more reliable and generally produce equal or higher predicted median and mean yields than medium maturity cultivars except in favoured areas .There was little evidence that medium maturity cultivars or higher plant populations offered any advantage except in favoured areas (Quirindi, Gunnedah), and in most cases variability in predicted yield was increased. It is clear that locality specific assessments using modeling are needed, as outcomes will differ from those for situations included here. Specifically, modeling supported by field studies for soils with greater water holding capacity (as occur in some of the Moree – Narrabri area and further west near Mungindi), even quicker maturity cultivar types and planting as early as late July in the warmer western areas (Mungindi) is needed. Because of warming temperatures and associated expected changes in rainfall, repeating the study using the most recent 30, 40 or 50 years weather data and with increased temperatures, as in Birch et al 2008b) would be informative. Also, the contribution by the individual factors explored here could be quantified by detailed statistical analyses of predictions. Nevertheless, the s study shows strategies to minimise risk of failure in more marginal areas include early and in some locations late planting of quick cultivar types. These approaches should optimize water use, minimise risk of high temperatures during sensitive stages of the crop growth and development and manage canopy size as for grain sorghum (Hammer 2006). Agronomic packages will also include land preparation practices that favour water accumulation, adequate nutrient supply and effective weed control.

References

Birch CJ, Robertson MJ, Humphreys E, Hutchins N. (2003). Agronomy of maize in Australia – in review and prospect. In Birch, C.J. and Wilson, S. (Eds) ‘Versatile Maize, Golden Opportunities’ Proc. 5th Australian Maize Conference, Toowoomba, 18-20 February 2003. Maize Association of Australia.

Birch CJ, Stephen K, McLean G, Doherty AG, Hammer GL Robertson MJ (2008). Assessment of reliability of short to mid season maize production in areas of variable rainfall in Queensland. Australian Journal of Experimental Agriculture 48: 326-334

Dimes JP, Revanuru S (2004) Evaluation of APSIM to simulate plant growth response to applications of organic and inorganic N and P on an alfisol and a vertosol in India. Tropical Soil Biology and Fertiliser Institure, Centro International de Agriculture Tropical (online) http://www.cgiarfinanceinfo.org/tsbf [Accessed 30/8/08]

Hammer GL (2006) Pathways to productivity: breaking the yield barrier in sorghum. Agricultural Science 19:16-22.

Keating BA, Carberry PS, Hammer GL, Probert ME, Robertson MJ, Holzworth D, Huth NI, Hargreaves JNG, Meinke H, Hochman Z, McLean G, Verburg K, Snow V, Dimes JP, Silburn M, Wang E, Brown S, Bristow KL, Asseng S, Chapman S, McCown RL, Freebairn DM, Smith CJ (2003) The Agricultural Production Systems Simulator (APSIM): its history and current capability. European Journal of Agronomy 18:267-288.

Lyon DJ, Hammer GL, McLean GB, Blumenthal M (2003) Simulation supplements field studies to determine no-till dryland corn population: Recommendations for Semiarid Western Nebraska. Agronomy Journal 95, 884–891.

Micheni, AN, Kihanda, FM , Warren, GP and Probert, ME (2004) Testing the APSIM model with experimental data from the long term manure experiment at Machang'a (Embu), Kenya. Tropical Soil Biology and Fertiliser Institure, Centro International de Agriculture Tropical (online) http://www.cgiarfinanceinfo.org/tsbf [Accessed 30/8/08]

O’Keefe K (2006) Maize: NSW planting guide 2006-07. NSW Department of Primary Industries, Sydney..

Robertson MJ, Cawthray S, Birch C, Bidstrup R, Crawford M, Dalgleish NP, Hammer GL (2003) Managing the risk of growing dryland maize in the northern region. In ‘Versatile Maize, Golden Opportunities’, Proceedings, 5th Australian Maize Conference. (Eds. CJ Birch, SR Wilson) pp 112-119 .(Maize Association of Australia, Darlington Point, NSW)

Routley, R and Robertson, M. J. 2003 Extending dryland maize production into environments of marginal moisture supply. Pp 180-186 in Birch CJ, Wilson SR (Eds.)‘Versatile Maize, Golden Opportunities’, Proceedings, 5th Australian Maize Conference (Maize Association of Australia, Darlington Point, NSW).

Previous PageTop Of PageNext Page