Previous PageTable Of ContentsNext Page

Plant populations to improve yield of dryland maize in northwest NSW

Sam Simons1, Daniel KY Tan1, Stephanie Belfield2 and Bob Martin3

1 Faculty of Agriculture, Food and Natural Resources, The University of Sydney, Sydney, NSW 2006, Australia. www.usyd.edu.au Email d.tan@usyd.edu.au

2 New South Wales Department of Primary Industries, Moree NSW 2400. www.dpi.nsw.gov.au Email stephbelfield@bigpond.com

3 New South Wales Department of Primary Industries, Primary Industries Innovation Centre, Tamworth NSW 2340. www.dpi.nsw.gov.au Email marti27@une.edu.au

Abstract

Maize is a crop that can contribute to the increasing need for both feed grains and silage in Australia. Rain-fed maize production has attracted much attention among farmers in the more marginal cropping regions of Australia as an alternative to traditional summer crop options. However, the optimal row spacing for dryland maize systems in northern NSW is still uncertain. Maize hybrids (Maximus, Hycorn 424), plant population density (10, 20, 30 and 50,000 plants/ha) and row configuration (solid row, single skip and double skip row) were evaluated at Moree, northwest NSW during the summer of 2006/07. Yield performance of the shorter season hybrid, Maximus was superior to Hycorn 424, achieving optimal yield at 20-30,000 plants/ha while the longer season hybrid, Hycorn 424 performed best at 10–20,000 plants/ha. However, above ground dry weight of Hycorn 424 was greater than that of Maximus, being positively correlated with increasing plant density, while dry weight of Maximus remained relatively stable with increasing plant density. It is possible that earlier maturing hybrids are less affected by severe terminal drought stress for grain production while later maturing hybrids are better for forage production. Single skip configurations achieved maximum grain yields and harvest index. The grain yield and harvest index levels were extremely low as consistent with a severe terminal stress during a below average rainfall season. This study highlights the riskiness associated with growing dryland maize in northwest NSW.

Key Words

dryland maize; plant population; row configuration; yield; harvest index

Introduction

Maize can contribute significantly to the increasing demand for both feed grains and silage in Australia. However, maize production in areas of marginal rainfall (<600 mm annual rainfall) is considered less reliable than grain sorghum (Robertson et al. 2003). High intra-seasonal variation in timing and amount of rainfall received in these marginal areas means that water stress is a major production limitation. Thus, maize has not found widespread acceptance as it is considered one of the riskiest crops to grow by many farmers (Robertson et al. 2003). Widely accepted strategies among dryland farmers to manage risk include fallowing, cultivar selection, optimising plant population density and planting date (Birch et al. 2008). Quicker maturing maize may be better able to take advantage of a full profile of soil water and minimise the risk of running out of water before the crop matures while avoiding the peak summer heat at tasselling when planted in spring. Plant population density and row configuration are two important cultural management decisions that influence crop growth, resource availability and yield potential. Some overseas work has been conducted investigating dryland maize populations (Westgate et al. 1997; Johnson et al. 1998; Norwood 2001; Lyon et al., 2003). In Australia, there has been substantial work done on optimising density and row configuration in dryland cotton (Bange et al. 2005) and sorghum (Whish et al. 2005) but not in dryland maize (Birch et al. 2008).

This paper evaluates the yield response of two maize cultivars differing in maturity (Maximus, Hycorn 424) growing in three row configurations (solid rows, single skip row, double skip row) in four plant population densities (1, 2, 3, 5 plants/m2) under rain-fed conditions in Moree, northwest New South Wales (NSW) during the summer of 2006/07.

Methods

Treatments and experimental site

The field experiment was conducted on a black vertosol at “Carossa” (29°5’S, 149°7’E) located approximately 25 km west of Moree in northwest NSW. The previous crop on the experimental site was wheat harvested in Dec 2005, and the stubble was burnt followed by a light cultivation. The maize was sown into soil with plant available water of approximately 80 mm on 1 Sep 2006, with 66 kg/ha N as urea and 40 kg/ha granular Starter Zn, and harvested on 19 Jan 2007. The precipitation received during the growing season was 190 mm, to the amount of 109 mm of effective rainfall compared with the long term rainfall average of 277 mm for Moree. This was a dry summer with most of the rain falling in small falls. Two maize (Zea mays) cultivars (Hycorn 424, Maximus) were planted in densities of 1, 2, 3 and 5 plants/m2 in solid, single skip and double skip row configurations with a row spacing of 0.92 m. Hycorn 424 is a medium maturing cultivar (115 CRM), while Maximus is a quick maturing cultivar (102 CRM). The experimental layout was a randomised complete block factorial design with three replicates.

Data collection

Volumetric water content was measured at 20 cm intervals to a depth of 160 cm on six occasions (26 Sep, 2, 10, 21 Nov, 1 Dec 06 and 16 Jan 07) during the growing season using a neutron probe, and plant available water calculated. Water use efficiency, defined as the crop grain yield (Y) per unit of total water use (TWU) was calculated as WUE = Y/TWU. TWU (mm) was approximated as TWU = P + rW, where P is precipitation (mm), and rW is change in soil water content from sowing to harvest (mm).

Date of maturity was estimated based on observation of black layer on at least 50% of plants in a plot. The number of cobs per plant and cobs per unit area were also recorded before harvest. Cobs from each biomass harvest sampling plot were handpicked in order to gain quality data and subsequently, hand threshed. The remaining cobs from the sample area were harvested by a small plot header. Grains were dried to 14% moisture and weighed. Harvest index was calculated for each plot as the proportion of grain weight (kg/ha) to the total above ground dry weight (kg/ha) at maturity. 100 grains were counted and weighed using electronic scales. Grains/cob, 100 grain weight measurements were based on 1 m2 subsamples (1 m length of 2 harvest rows), whereas cobs/ha, grain yield and harvest index were based on whole plot harvests. The data was analysed by analysis of variance (ANOVA) and regression using Genstat v10.0.

Results and discussion

Analyses of variance (Table 1) indicated significant cultivar by density interactions and row configuration effects.

Cultivar by density interaction

Grain yield in the later-maturing Hycorn 424 declined linearly with increasing density (R2=0.94, P<0.05, Fig. 1a), while yield in Maximus peaked at 20-30,000 plants/ha, and declined at 50,000 plants/ha (R2=0.90, P<0.05, Fig 1a). The decreased yield at higher densities was primarily due to the reduction in cobs/ha, and to a lesser extent, the reduction in grains/cob and 100 grain weight (data not presented) with increasing densities. As this study was conducted in a severe terminal stress year, our grain yields (< 1 t/ha) were achieved at lower plant populations than in studies in the USA where maximum grain yield was achieved at plant densities of 54,000 plants/ha (Norwood 2001) and 100,000 plants/ha (Johnson et al. 1998). Under favourable rainfall conditions in Kenya, 70,000 plants/ha resulted in maximum grain yield, while 20,000 plants/ha achieved optimal yield in less favourable years (Nadar 1984).

Our cultivar by density interaction was consistent with recent simulation studies for Moree that predicted grain yields of 2-3 t/ha for a quick maturing hybrid at low populations of 20,000 plants/ha, while medium maturing hybrids at 20-40,000 plants/ha would yield < 2 t/ha (Birch et al. 2008). Yields in our study rarely approached the 2 t/ha simulated in an average year, and yield penalties occurred at plant densities >10,000 plants/ha in the later-maturing Hycorn 424. The unusually dry finish in the season of our experiment would place it at the lower end of the distribution of the simulation study by Birch et al. (2008).

Hycorn 424 had higher above ground dry weight than Maximus. Harvest index of both Hycorn 424 and Maximus declined linearly as plant density increased. Above ground dry weight in Hycorn 424 increased linearly with increasing plant density (Fig. 1b), but did not consistently increase with plant density in Maximus (Fig. 1b). Hence, above ground dry weight had a negative relationship (y = -0.06x + 902, R2=0.28, P<0.05) with grain yield.

Table. 1. F-test level of significance to determine the effect of cultivar, row configuration and plant density on grain yield, above ground dry weight, harvest index, water use efficiency (WUE), cobs/ha, grains/cob and 100 grain weight under dryland conditions in Moree, northwest NSW.

Variables

Cultivar

Row config

Density

Density*row config

cultivar*density

Row config*density

Cultivar*Row config*density

Grain yield (kg/ha)

***

***

***

n.s

*

n.s

n.s

Above ground DW (kg/ha)

***

n.s

***

n.s

***

n.s

n.s

Harvest index

***

**

***

n.s

*

n.s

n.s

WUE (grain) (kg/ha/mm)

n.s.

n.s.

*

n.s.

n.s.

n.s.

n.s.

Cobs/ha

***

***

n.s

***

***

n.s

*

Grains/cob

n.s

n.s

***

n.s

n.s

n.s

n.s

100 grain weight (g)

n.s

n.s

***

*

n.s

n.s

n.s

* significant at the P<0.05, ** significant at P<0.01, *** significant at P<0.001 level, n.s – not significant at P=0.05.

Figure 1. The effect of plant density and cultivar on maize (a) grain yield (kg/ha) and (b) above ground dry weight (kg/ha) in Hycorn 424 (®) and Maximus (¾). Vertical bar represents l.s.d at P=0.05. Data presented are cultivar x plant density interaction means (n = 9).

Our yield and harvest index levels were extremely low as consistent with severe terminal stress. The later-maturing Hycorn 424 was able to access far more water, possibly due to increased TWU and almost doubled its above ground biomass at higher density but was unable to convert this to yield. Hence, Hycorn 424 used up available water prior to anthesis and suffered flowering and terminal stress. This supports other research showing that high vegetative organ reserves at anthesis did not necessarily increase stress tolerance in hybrids and ultimately grain yield (Monneverux et al. 2005). Our study highlights the riskiness associated with growing dryland maize in northwest NSW and is consistent with the findings of Birch et al. (2008). During severe stress years with low grain yields, it may be possible harvest above ground biomass (crop stover or residue) to supply hay and silage to feedlots. Furthermore, maize above ground biomass may also have potential as cellulosic feedstock for the next generation of ethanol biofuel production, although this technology is still too expensive to be commercially available at the moment (Odeh and Tan 2007).

Effect of row configuration

Single skip row produced higher (P<0.05, l.s.d = 92.3 kg/ha) mean grain yield (655 kg/ha) than solid rows (510 kg/ha) and double skip (507 kg/ha). Single skip row (0.13) also had a higher (P<0.05, l.s.d = 0.019) mean harvest index than solid row (0.10) but was not higher (P>0.05) than double skip row (0.11). The double skip configuration did not appear to allow sufficient water to be conserved in the double skip row for use during the grain filling period, which is similar to a previous study on skip row sorghum (Thomas and Myers 1981). It is possible that double skip configuration was too wide for the maize to extract water from the centre of the double skip although we have no data to support this.

Conclusion

Grain yield of fast maturing hybrid, Maximus was higher than Hycorn 424, achieving optimal yield at 20-30,000 plants/ha while Hycorn 424 performed better at 10-20,000 plants/ha. Hycorn 424 above ground dry weight was greater than that of Maximus. The grain yields and harvest index levels were extremely low in this study as consistent with severe terminal stress during a dry year. Based on one season’s field data, earlier maturing cultivars such as Maximus appear to be less affected by severe terminal drought for grain production while there is potential for silage and hay production for late maturing cultivars such as Hycorn 424. These are preliminary findings in a season with lower than average rainfall. Long term field experiments and simulations need to be conducted to evaluate the sustainability of dryland maize production for northwest NSW.

Acknowledgements

Partial financial support by ACIAR (Australian Centre for International Agricultural Research, Project ID: ASEM/2000/109 – Farming systems for crop diversification in Cambodia and Australia) is gratefully acknowledged.

References

Bange MP, Carberry PS, Marshall J, Milroy SP (2005). Row configuration as a tool for managing rain-fed cotton systems: review and simulation analysis. Australian Journal of Experimental Agriculture 45, 65-77.

Birch CJ, Stephen K, McLean G, Doherty A, Hammer GL, Robertson MJ (2008). Reliability of production of quick to medium maturity maize in areas of variable rainfall in north-east Australia. Australian Journal of Experimental Agriculture 48, 326-334.

Johnson GA, Hoverstad TR, Greenwald RE (1998). Integrated weed management using narrow row corn spacings, herbicides and cultivation. Agronomy Journal 90, 40-46.

Lyon DJ, Hammer GL, McLean GB, Blumenthal JM (2003). Simulation supplements field studies to determine no-till dryland corn population recommendations for semiarid western Nebraska. Agronomy Journal 95, 884-891.

Monneveux P, Zaidi H, Sanchez C (2005). Population density and low nitrogen affects yield associated traits in tropical maize. Crop Science 45, 535-545.

Nadar HM (1984). Maize yield response to row spacings and population densities under different environmental conditions. East African Agricultural and Forestry Journal 43, 157-164.

Norwood CA (2001). Dryland corn in western Kansas: Effects of hybrid maturity, planting date, and plant population. Agronomy Journal 93, 540-547.

Odeh IOA, Tan DKY (2007). Expanding biofuel production in Australia: Opportunities beyond the horizon. Farm Policy Journal 4, 29-39.

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

Thomas GA, Myers RJ (1981). Evaluation of row spacing and population density effects on grain sorghum over a range of northern Australian environments. Australian Journal of Experimental Agriculture 21, 210-217.

Westgate ME, Forcella F, Reicosky DC, Somsen J (1997). Rapid canopy closure for maize production in the northern US corn belt: Radiation-use efficiency and grain yield. Field Crops Research 49, 249-258.

Whish J, Butler G, Castor M, Cawthray S, Broad I, Carberry P, Hammer G, McLean G, Routley R, Yates S (2005). Modelling the effects of row configuration on sorghum yield reliability in north-eastern Australia. Australian Journal of Agricultural Research 55, 11-23.

Previous PageTop Of PageNext Page