Abstract

Media summary

Key Words

Subsoil constraints, salinity, sodicity, crop lower limit, plant available water capacity

Introduction

Successful dryland crop production on the clay soils of southwest Queensland is dependent on the utilisation of soil moisture accumulated in the preceding fallow period (Freebairn et al. 1990). High levels of subsoil constraints, particularly salinity and sodicity, are known to be present in many soils of the region (Dalal et al. 2002; Routley 2003). These constraints are believed to restrict rooting depth and increase lower limit of plant available water (Crop Lower Limit, CLL) and hence the ability of crops to utilise stored soil moisture (Dalal et al. 2002).

Options to ameliorate these subsoil constraints appear to be limited and management solutions may be limited to the understanding the impact of the constraints on the performance of various crop options so that appropriate agronomic decisions can be made (Dang et al. 2004). The variable distribution of subsoil constraints, both spatially across the landscape, and with depth in the soil profile, and the complex interactions that exist between the constraints, make it difficult to derive simple relationships between constraints and crop yield. Knowledge of the effect of subsoil constraints on factors influencing yield, such as CLL, will assist in the development of decision support tools (including systems modelling) that will ultimately allow producers and advisors to make informed decisions about managing production systems where subsoil constraints are a limiting factor.

The objectives of this study were to (i) quantify the relative tolerance of a range of common winter crops to subsoil constraints, and (ii) quantify the relationships between CLL and subsoil constraints.

Materials and Methods

Eight field experiments were established on farms in southwest Queensland in the winter cropping season of 2003, on soils with a range of sub-soil constraints. Treatments consisted of various combinations of different crop species (durum wheat cv. Yallaroi, bread wheat cv. Baxter, barley cv. Mackay, canola cv. Hyola 43 and chickpea cv. Jimbour). All experiments were planted in a complete randomised block design with 3 replications. All planting, harvest and crop management operations were carried out using the co-operating farmers equipment and planting rates and other management practices followed the accepted district practice. All crops were well managed with no significant weeds, pests, diseases or nutrient deficiencies experienced.

Soil water was monitored (0-130 cm) throughout the season by either neutron moisture meter or coring for gravimetric soil water at sowing and at physiological maturity. The soils at all sites were characterised for bulk density, DUL, CLL and PAWC (PAWC = DUL-CLL) using the method of Dalgliesh and Foale (1998). The effect of crop species on various components of the soil water balance was analysed using Restricted Maximum Likelihood (REML) variance analysis on Genstat 6.1.

Soils at each site were analysed for pH, EC, Cl (1:5 soil: water extracts), clay content, and cations after alcoholic displacement (Rayment and Higginson 1992) in 0-10 cm and 20 cm depth increments thereafter. ECse was calculated from EC (1:5 soil:water extracts), Cl and clay content using the method of Shaw (1999).

Regression analysis was used to establish relationships between soil characteristics and CLL using 174 data points (crop x site x soil depths). Threshold levels of subsoil constraints were determined using a two-step linear regression analysis approach: (i) the effect of soil texture on CLL was accounted for, and (ii) the residuals of CLL were regressed against subsoil constraints (Sadras et al. 2003). The chemical constraints having a significant positive correlation with residuals of CLL were used to obtain threshold levels calculated as (-intercept/slope). The top layer (0-0.1 m) was excluded from the analysis to avoid confounding effects of evaporation and plant water uptake on minimum soil water content (Sadras et al. 2003).

Results and Discussion

Soil characterization

Soil profile distribution of pH, electrical conductivity (ECse), chloride concentration (Cl), and exchangeable sodium percent (ESP) at the experimental sites are given in Figure 1. The dominant soils at the sites were grey, brown and red Vertosols and Sodosols.

Figure 1. Soil profile distribution of pH, ECse, chloride and ESP of the trial sites.

Crop species

Table 1 compares the PAWC at several sites with a range of subsoil constraints, for 5 crops, for each depth increment in the soil profile. At P ≤0.05, the total PAWC for chickpea was significantly lower than for the canola and bread wheat due to a reduction in its ability to extract water from soil layers within its rooting zone. The total PAWC for canola was significantly higher than for all other crops. PAWC for durum wheat was significantly lower at all depths as compared to bread wheat, barley and canola.

Table 1. Means of PAWC (mm) of soils with range of subsoil constraints, for 5 crop species. Number in parenthesis indicates the number of sites at which PAWC was determined for that crop. PAWC values within a row, followed by same letter do not differ significantly at P<0.05.

Depth (cm)	Durum wheat (4)	Bread wheat (8)	Barley (6)	Chickpea (4)	Canola (4)
10-30	39.7a	49.9bc	46.9b	47.3b	55.3c
30-50	30.1a	37.4bc	34.9b	29.3a	40.7c
50-70	25.1b	29.0b	28.3b	17.7a	30.7b
70-90	18.6a	23.4b	22.9b	17.4a	27.4b
90-110	9.4a	14.1b	13.6b	13.7b	16.5b
110-130	2.9a	10.5b	10.2b	15.5c	19.9d
Total	125a	164c	156bc	141ab	190d

The low PAWC for durum wheat could be due to significantly higher Na⁺ accumulation and significantly low selectivity for K⁺ in the leaves (Table 2). Munns et al. (2000) reported that sensitivity of durum wheat to salts is due to the fact that it lacks the genes for both the ability to restrict Na⁺ uptake as in bread wheat and to tolerate high Na⁺ levels as in barley.

Although the selected sites had no evidence of acidity except for site 1 where pH drops to <5.0 at 1 m, all the crops accumulated aluminium (Al), ranging from 20-230 mg/kg in the youngest mature leaf (YML). Recently, Ma et al. (2003) reported the toxicity of anionic aluminate ions [Al(OH)₄^-] in alkaline pH. On an average, chickpea accumulated significantly higher concentrations of Al in the YML than other crops. Chickpea also accumulated significantly higher concentrations of B in the YML than other crops. This high B concentration was above the critical toxic concentration of B for chickpea (Chapman 1996). It is unclear if this high accumulation of Al and B in the YML of chickpea could result in low PAWC and low soil water extraction.

Table 2. Nutrients concentration in youngest mature leaf

Crop	KA (%)	NaB (%)	AlC (mg/kg)	BD (mg/kg)
Wheat cv. Yallaroi	1.57a	0.39b	19a	8.8a
Wheat cv. Baxter	2.58c	0.01a	56a	18.1b
Barley cv. Mackay	2.79cd	0.33b	69a	17.9b
Canola cv. Hyola 43	2.33bc	0.57c	130b	26.3c
Chickpea cv. Jimbour	2.05b	0.02a	230c	38.6d

^{A, B,C,D}Means followed by same letter within a column do not differ significantly at P<0.05.

Relationships between CLL and subsoil constraints

Clay content accounted for 21% (P < 0.001) of the variation in CLL. Significant positive relationships between CLL residuals and ECse, ESP and Cl were found. Table 2 provides the parameters of the linear regression between residuals of CLL and subsoil constraints after accounting for the effect of texture. The calculated threshold values suggested that CLL is increased when subsoils have an ECse >8.5 dS/m, ESP >17.0 and Cl >672 mg/kg.

Table 3. Parameters of linear regression between crops lower limit after accounting for the effect of texture (clay %) and subsoil constraints in southwest Queensland

Subsoil constraints	Parameters
	Intercept (S.E.)	Slope (S.E.)	R2	Threshold*
ECse (dS/m) ESP (%) Chloride (mg/kg)	-0.495 (0.11) -1.737 (0.21) -0.519 (0.10)	0.058 (0.01) 0.1021 (0.01) 0.00077 (0.0001)	0.15*** 0.30*** 0.22***	8.53 17.0 674

* P <0.05, ^** P < 0.01, ^*** P < 0.001

After accounting for the effect of texture, a step-wise multiple regression model explained 41% of the variation in CLL (Eq. 1).

Y_{(residual CLL)} = 0.0948 ESP + 0.049 ECse – 2.027, r² = 0.41, P < 0.001 (Eq. 1).

Threshold values were also obtained for individual crops after accounting for the effect of texture (Table 3). The results suggested a range of threshold values for different crops. Durum wheat was more sensitive to subsoil salinity than bread wheat and barley while bread wheat and barley were more sensitive to subsoil sodicity as compared to durum. Chickpea was found to be most sensitive to subsoil sodicity.

Table 4. Threshold calculated from the parameters of linear regression between crop lower limit after accounting for the effect of texture and subsoil constraints

Crop	ECse (dS/m)	ESP (%)	Cl (mg/kg)
Wheat Yallaroi Wheat Baxter Barley Mackay Chickpea Jimbour Canola Hyola 43	4.69 8.78 8.31 ns ns	18.3 16.6 17.2 14.8 ns	506 656 695 ns ns

ns, non significant

Results provided an early indication of threshold values as the analysis included only a small number of crops grown on soils with narrow ranges in subsoil constraints.

Acknowledgements

The Grains R&D Corporation funded this research. The generous support of our collaborative growers and their families in providing sites and managing the trials is greatly appreciated. Thanks are also due to Kerry Bell, Biometrician for providing help with statistical analysis.

References

Chapman VJ (1996). Boron nutrition of winter crops with particular reference to boron toxicity. PhD thesis, University of Queensland, Brisbane.

Dalal RC, Blasi M, So HB (2002). High sodium levels in subsoil limits yields and water use in marginal cropping areas. Final report GRDC project no. DNR 6.

Dang YP, Dalal RC, Routley R, Schwenke GD, Daniells I (2004). Subsoil constraints to grain production in the cropping soils of the north-eastern region of Australia. Australian Journal of Experimental Agriculture (submitted).

Dalgleish NP, Foale MA (1998). Soil Matters: Monitoring soil water and nutrients in dryland farming systems. (Agricultural Production systems Research Unit, Toowoomba, Queensland).

Freebairn DM, Littleboy M, Smith GD, Coughlan KJ (1990). Optimising soil surface management in response to climatic risk. In ‘Climatic risk in crop production: models and management for semiarid tropics and subtropics (Eds. R.C. Muchow, J.A. Bellamy). pp. 283-305. (CAB International Wallingford, UK).

Ma G, Rengasamy P, Rathjen AJ (2003). Phytotoxicity of aluminium to wheat plants in high-pH solutions. Australian Journal of Experimental Agriculture 43, 497-501.

Munns R, Hare RA, Colmer TD (2000). Salt tolerance in wheat and barley. In ‘Proceeding 8^th International Barley Genetics Symposium 22-27 Oct. 2000, Adelaide (Ed. S. Logue) Vol.1: 238-245. (University of Adelaide, South Australia).

Rayment GE, Higginson FR (1992). Australian laboratory handbook of soil and water chemical methods. (Inkata Press: Melbourne).

Routley R (2003). Sub-soil constraints in the Maranoa-soil survey and crop monitoring results from 2002. Roma Grains Research Update 2003. 17 March 2003, pp. 19-22.

Sadras V, Beldock J, Roget D, Rodriguez D (2003). Measuring and modelling yield and water budget components of wheat crops in coarse-textured soils with chemical constraints. Field Crops Research 84: 241-260.

Shaw RJ (1999). Soil salinity-electrical conductivity and chloride. In ‘Soil analysis-an interpretation manual (Eds. K.I. Peverill, L.A. Sparrow, D.J. Reuter). pp. 129-145. (CSIRO Melbourne).