Key messages

• Surface application of at least 2t/ha lime will increase yield within 1–8 years of application, however, higher lime rates will maintain and improve subsoil pH over time and might increase yield further.
• Economic breakeven can be achieved after five years of cropping where the initial topsoil pH is greater than 5.0. However, if the topsoil pH is 4.6 or lower, economic benefit will occur almost immediately.
• Where the subsoil is acidic and compacted, incorporation of lime using strategic deep tillage is warranted.
• It is recommended to use the most readily available and cheapest test to diagnose subsoil acidity. In most cases this will be soil pH.

Aims

To evaluate the effectiveness of applying lime to manage subsoil acidity using various soil amelioration approaches. To estimate the rate of lime required to ameliorate subsoil acidity under minimum and strategic tillage. To evaluate the economic benefit of applying lime using different soil amelioration approaches. To compare measurement of soil pH and extractable aluminium as indicators of the likely effects of subsoil acidity.

Introduction

About 50% of the subsoil of the Western Australian (WA) wheatbelt is constrained by severe acidity (low pH and aluminium toxicity) to a deeper depth than any other crop growing regions of Australia (Gazey et al 2013; Li et al 2019). More importantly, in WA soil acidity often coincides with other soil constraints such as compaction and topsoil water repellence (van Gool 2016). In soil with multiple constraints, most crop roots are confined to within 20–25cm of the soil surface (Azam and Gazey 2020), allowing a large proportion of growing season rainfall to leach rapidly beyond the root zone, which results in ineffective use of subsoil moisture (Azam et al 2019a).

In WA, several soil amelioration approaches are available for growers to manage subsoil acidity. The typical approach is to spread agricultural lime on the soil surface (often involving multiple applications). This approach takes many years to increase subsoil pH, but adequate surface liming can neutralise further acidification of the subsoil (Azam and Gazey 2020). Given the slow movement of alkali below the depth of lime addition, an alternative opportunistic and quicker approach is to use physical tillage operations, whereby lime is spread and incorporated while loosening compacted soil by a deep ripper or overcoming water-repellent soil by inversion ploughing (Davies et al 2019). Incorporation of limed topsoil is another suggested approach, for farmers who have repeatedly applied lime and built a healthy topsoil with pH 6.5–7.5. In this approach the amount of undissolved lime is estimated to identify a potential depth of physical incorporation of the limed soil (Azam and Gazey 2020). There is also soil profile re-engineering to address multiple constraints to a deeper depth. This approach is currently difficult (expensive) to adopt at a farm scale but is useful as a means of gauging the grain yield potential at a particular site in the absence of soil constraints (Azam and Gazey 2019). The latter three approaches require the re-introduction of tillage to a 20-plus year no-till cropping system, which may have a minor short-term negative effect on soil health, (e.g. soil structure and organic carbon levels) especially when shallow incorporation is carried out (Kirkegaard et al 2020).

Recent laboratory and long-term field experiments have been conducted to examine the impact of these approaches on soil acidity and crop production. This paper presents a combined analysis of the results from these experiments to determine:

1. how often and how much lime to apply under amelioration approaches.
2. the economic implications of different soil amelioration treatments.
3. the best way to measure the impact of liming on soil health and root growth.

Method

Short- and long-term lime experiments (with or without strategic tillage)

The Department of Primary Industries and Regional Development (DPIRD) has conducted several short- and long-term soil acidity management experiments in WA, in which a wide range of lime rates (cumulative total 0–8.5 t/ha) was applied to the soil surface on small plots (width 2–4m, length 20–30m).

In a long-term experiment conducted at Wongan Hills (30°85’S, 116°74’E), five rates of lime (90% neutralising value, 99% of particles <0.5mm) were applied in 1994. The plots were subdivided in 1998 and 2014 for application of additional lime rates to the soil surface (full method detailed in Azam et al 2019b). All plots were ripped to 50cm in 2016. In 2018, three strategic tillage treatments were applied to incorporate residual lime. Lupin, wheat, canola and barley crops were grown in an irregular sequence over 27 years.

In an experiment at Merredin (31°29'S, 118°13'E), 0, 2, 4 and 6t/ha rates of lime (94.9% neutralising value, 99% of particles <0.5mm) were applied in 2017 with and without shallow, strategic tillage (offset plough, 10–15cm deep). The whole site was ripped to 50cm before lime application. Wheat was grown continuously for five years.

At Kalannie (30°25’S, 117°17'E), lime (91.9% neutralising value, 99% of particles <0.5mm) was applied at the same rates as in the Merredin experiment either to the surface or incorporated to a depth of 20cm using a one-way plough (full method detailed in Azam et al 2019c). The entire site was ripped to 50cm before lime application. A wheat-wheat-canola-barley crop sequence was grown over the following four years.

At Tardun (28°71’S, 115°91E), 0, 2 and 4t/ha rates of lime (94% neutralising value, 99% of particles <0.5mm) were initially applied in 2013 with three incorporation treatments (nil, 10–15cm and 20-25cm). The plots were subdivided in 2019 and an additional 2t/ha lime was applied to the surface of half of the plots (full method detailed in Reynolds et al 2021). Wheat was grown each year between 2013–2020, except for a chemical fallow in 2018.

At West Casuarinas (28°95’S, 115°26’E), an experiment was established in 2007 in which 0 or 2t/ha lime (94% neutralising value, 99% of particles <0.5mm) was applied either to the soil surface or incorporated with a mouldboard plough to a depth of 25–30cm (full method detailed in Davies et al 2015). Crops were grown continuously for 13 years using a wheat-lupin-barley-lupin-wheat rotation. All plots were subdivided in 2017 and an additional 3t/ha lime (94% neutralising value, 99% of particles <0.5mm) was applied to the soil surface on half of each plot.

A second, deep lime incorporation (profile re-engineering) experiment was established at Kalannie (30°25’S, 117°17'E) where 0, 1.5, 4.5 and 6.0t/ha rates of lime (94.9% neutralising value, 99% of particles <0.5mm) were used (full method detailed in Azam and Gazey 2019). The experiment was established in 2018 and monitored for three seasons. There were five soil amelioration treatments comprising an untreated control, removal of compaction only and removal of both compaction and acidity (by complete incorporation of lime at 0–10, 0–30 and 0–45cm).

From all experiments, soil profile samples were collected at 0–10, 10–20, 20–30, 30–40 and 40–50cm depths from each plot to measure soil pH and Al concentration using 1:5 soil:0.01M CaCl₂ extract. Crops were harvested using a plot harvester for measuring yield, except for the experiment at Tardun and the re-engineering experiment. At Tardun, yield was determined from the farmer’s yield monitor in the first five years, hand cuts in 2019 and small plot harvester in 2020. In the re-engineering experiment at Kalannie, crops were harvested by hand-cuts and the moisture content of the soil profile was measured in situ regularly using capacitance probes. Crop root architecture was also imaged repeatedly in situ using a Rhizotron set up (see details in Azam and Gazey 2020).

Glasshouse experiment assessing sensitivity of cereal roots to subsoil pH and aluminium

A large glasshouse experiment was conducted at the DPIRD Northam facility (31°65’S, 116°70'E). Bulk acidic subsoil (20–40cm depth) with different pH and extractable aluminium levels was collected from Wongan Hills (30°25’S, 117°17'E), Kalannie (30°25’S, 117°17'E) and Northampton (28°09’S, 114°59’E). Another set of benign subsoil (below 50cm depth) that was less acidic and had no extractable aluminium was collected from Wongan Hills and Northampton. The two sets of subsoil from Wongan Hills and Northampton were then mixed, separately for each location, without any lime, at different ratios to achieve five pH (and aluminium) levels. The Kalannie acidic subsoil was mixed with Wongan Hills benign subsoil (as the Kalannie deep subsoil was acidic but had similar clay content to the Wongan Hills site) to achieve six pH (and aluminium) levels. Four kilograms of mixed soil was filled in a tapered pot, wet to field capacity, incubated for seven days and then planted with two wheat and two barley varieties (details to be published in a journal article). Crops were grown in the pots in a controlled environment glasshouse for eight weeks before they were harvested. About 200g soil sample was collected from each pot to measure soil pH and Al concentration in 0.01M CaCl₂. Roots were separated from soil by washing, scanned and root length was measured using WinRHIZO (v. 2005c; Regent Instruments Inc., Quebec City, Canada) software.

A linear model was fitted to each of the measurements using the ANOVA procedure in GenStat (Version 18.1, VSN International, Oxford, UK) to compare the treatment effects on soil pH, aluminium, root growth and crop yield. Fisher's protected least significant difference (LSD) was applied at the 0.05 significance level.

Economic analysis of long-term experiments

The cumulative net benefit of lime application to the soil surface with and without incorporation was estimated for the four experiments (Merredin, Kalannie, West Casuarinas and Wongan Hills). Information from each experiment was used in the analyses (Table 1). The benefit (or loss) in each year was determined from the product of the yield difference between lime treatment and control and the value of the crop. The cost was determined from the total cost of lime (product and freight plus spreading and/or incorporation). The benefit (or loss) each year was cumulated over the duration of the trial. For the Wongan Hills experiment, an app ‘iLime’ (https://www.agric.wa.gov.au/apps/ilime) was used to estimate cumulative benefit over 27 years, since yield had only been measured in eight of the years. The app was initialised using a sandy earth soil with the pH of the paddock before treatment, the cost of lime in 1994, the maximum yield of the control in each sampled year (1994, 1995, 1998, 2000, 2012, 2018–2020) or, in the case of years for which no data were collected, of a wheat crop grown in the same paddock in 1994. Default values from the app were used for all other inputs. Prices and costs were not discounted in the analyses of Merredin, Kalannie, West Casuarinas, but were discounted (7%) for the longer-term analysis of Wongan Hills using iLime.

Results

Surface application of lime

The time to gain significant yield increases from the surface application of lime depended on the initial topsoil pH and the rate of lime application (Figure 1). The initial pH of the topsoil at Wongan Hills and West Casuarinas was greater than 5.0. At Wongan Hills, it took seven years to gain any significant yield response (Figure 1a) while at West Casuarinas the first significant effect on yield was found in the eighth season (data not presented). Topsoil in the short-term experiments at Merredin (Figure 1b), Kalannie (Figure 1c) and Tardun (see Reynolds et al 2021) had initial pH less than 4.6. Yield increased significantly from the first season at Merredin and Kalannie and by the third season at Tardun.

In the Wongan Hills experiment, the yield response in the seventh year was significant only with the 4t/ha lime treatment. By 2012, (19 years after the first application of lime and the next time a crop was measured), surface application of both 2 and 4t/ha lime generated significantly greater yield for wheat. The same was found for wheat, canola and barley in every year from 2018–2020. Rates of lime below 2t/ha never yielded significantly more than the control. No significant differences in crop yield were observed between 2 and 4t/ha of lime.

In the experiment at Merredin (Figure 1b), Kalannie (Figure 1c) and Tardun (Reynolds et al 2021), application of 2, 4 and 6t/ha lime to the soil surface significantly increased yield from 1–3 years after application and continued to yield higher, compared to the control. There was no significant yield difference between the lime rates, but in later years, higher lime rates showed a non-significant trend of yielding more compared to 2t/ha lime treatment (Figures 1b & 1c).

Figure 1: Grain yield responses due to surface application of lime in (a) Wongan Hills, (b) Merredin and (c) Kalannie experiments. * indicate significant yield increases over the 0t/ha lime at P=0.05.

At Wongan Hills, only cumulative higher lime rates (>6.5t/ha from multiple applications) increased subsoil pH in the 10–20cm depth to the minimum target level (see Azam and Gazey 2020). There was no effect at depths below 20cm compared to the initial pH measured in 1994. However, higher rates of lime always had higher pH in 0–10 and 10–20cm depths and some lime remained undissolved in the top 4 cm surface soil (see Azam and Gazey 2020). This undissolved lime would neutralise ongoing acidification, maintaining soil pH and crop yield potential. In contrast, in the short-term experiments (Merredin, Kalannie and Tardun), none of the lime rates significantly increased subsoil pH below 20cm depth over the 4-9 year duration (see Azam et al 2019c and Reynolds et al 2021).

Incorporation of new and residual lime

Shallow incorporation treatments (10–20cm deep) did not increase grain yield over surface applied lime within 4–7 years in any of the three short-term experiments (see Azam et al 2019c and Reynolds et al 2021). However, in the West Casuarinas experiment, incorporation of 2t/ha lime using a mouldboard plough (MBP) to a depth of 25–30cm significantly increased yield after seven years and has continued to have a yield advantage until now, 14-seasons after application, compared to the MBP-alone and the control (Figure 2). At the West Casuarina experiment, incorporation of 2t/ha lime significantly increased soil pH to the depth of incorporation, however, topsoil pH did not increase to the minimum target pH of 5.5 (see Davies et al 2015). At Wongan Hills, incorporation of residual lime to 25cm using a rotary hoe rapidly increased pH above the minimum target level (pH 4.8) and significantly increased wheat yield in the year of treatment (see Azam et al 2019b). However, the effect on the yield of canola in 2019 and barley in 2020 were not significant (data not presented).

Re-engineering acidic compacted soil

In the re-engineering experiment at Kalannie, the combined removal of compaction and acidity at least doubled the yield of wheat, canola and barley (Figure 3). Removal of compaction alone increased yield of wheat but not of canola or barley. Grain yield exceeded the water-limited potential (as calculated using the French and Schultz (1984) equation) by 33─56% due to amelioration of multiple constraints under standard agronomic practice. Deep soil re-engineering allowed plants to produce root systems 60–65cm deep, while they were only 20–25cm deep for the control (data to be presented at the GRDC Research Update 2021, Perth). Significant and rapid improvement in root growth and yield were achieved due to increase in subsurface soil pH as well as a uniform and sustained decrease in soil resistance (see Azam and Gazey 2020).

Economic implications

The cumulative net benefit, in comparison to the control, shows that application of lime to the soil surface generated economic benefit within 1–7 years, depending on the rate of lime applied and the pH of the soil (Figure 4).

In the Wongan Hills experiment, modelled results from iLime showed that treatment with lime did not break even (compared to the control) until five years after the original application (Figure 4a). After this break-even year, the return for the lime treatments was greater than the control, year on year. From the fifth to the 17th year (1998–2010), the cumulative net benefit was equivalent for all applications of lime (Figure 4a, only the initial 2t/ha and 4t/ha treatments have been shown for clarity). Applications of 3.5t/ha lime or greater resulted in a higher cumulative net benefit for the final ten years of the experiment. This was due to a decline in the pH of the 0–10cm and 10–20cm soil layers for the 2t/ha treatment compared with the higher rates of lime (data not shown).

Figure 4. Cumulative net benefit ($AU/ha) in (a) Wongan Hills, (b) West Casuarinas, (c) Merredin and (d) Kalannie experiments. Net benefit was calculated as the sum of the return in each year (yield x price for that year), less the return for the control in that year and the total cost of each treatment. Yield from the experiments was used, except for Wongan Hills, which was estimated from the iLime app. Prices and costs were discounted by 7% for the 27-year analysis at Wongan Hills but were not discounted for the shorter analyses at the other locations.

In the West Casuarinas experiment, mouldboard plough (MBP) on its own increased crop yield response, and thus net returns, from the fourth season to the ninth season, but after that the crop response, and hence returns, plateaued (Figure 4b). The combination of MBP and lime became more profitable than MBP only from the seventh season onwards and profitability has increased since then to the present.

The breakeven point was reached a year after the application of lime at Merredin (Figure 4c) and Tardun (Reynolds et al 2021) while it took two years to reach breakeven at Kalannie (Figure 4d) and only for the 2t/ha treatment. In all three short-term experiments, higher lime rates initially lagged behind in net economic benefit compared to the 2t/ha lime due to the higher cost of treatment. Towards the end of this reporting period, the cumulative benefit from higher rates of lime was close to those for 2t/ha, emphasising that these higher rates might neutralise ongoing acidification maintaining soil pH, crop yield potential and hence the economic benefit.

Diagnostics of acidic subsoil

In constructed subsoil, the pH and extractable aluminium levels measured in 0.01 CaCl₂ varied between locations (Figure 5). Soil from Kalannie had significantly higher aluminium values at the same pH than soil from either Wongan Hills or Northampton. Soil pH and aluminium levels were strongly related to each other within each soil type and were equally related to growth of crop roots (data not shown, will be presented at the 2021 Research Updates). At soil pH of 4.0 and below the extractable aluminium reached toxic levels for each soil (Figure 5).

Figure 5: Soil pH and aluminium relationships for different soil types in 0.01M CaCl₂ extract. Vertical error bars represent the standard error of the mean values.

Conclusion

Surface lime application, in two long-term experiments, took 7–8 years to generate a significant increase in crop yield. The economic breakeven was achieved after five years of cropping at both sites with an initial topsoil pH greater than 5.0. However, in three short-term experiments, where the topsoil pH was 4.6 or less, crop yield and economic benefit occurred almost immediately. Once yield had increased in the lime-treated plots, in most experiments, they continued to yield more and increase in economic benefit year on year, compared to the un-limed control and irrespective of crop type. In all experiments, surface application of 2t/ha lime increased yield to an optimum level. However, lime rates higher than 2t/ha resulted in comparatively higher pH and more undissolved lime in the surface soil, and therefore might have the potential to increase yield in the future and to neutralise ongoing subsoil acidification.

Incorporation of lime using strategic tillage was effective in increasing crop yield where the depth of incorporation was 25cm or deeper. The response in yield was greater from incorporation to deeper depths and with higher lime rates than for shallower depths with lower lime rates, with the deepest lime incorporation treatment at Kalannie lifting the grain yield above the calculated water-limited yield potential. Therefore, we recommend incorporation of lime using strategic deep tillage for compacted soil with subsoil acidity.

It is essential to diagnose acidic subsoil using a reliable and cost-effective method. Our results confirm that soil pH and extractable aluminium values are strongly correlated and measuring one soil parameter will suffice as a diagnostic of subsoil acidity.

References

Azam G, Gazey C, Bowles R (2019a) Dynamics of water use by wheat and canola crops in compacted, acidic sands treated with deep strategic tillage and lime. In ‘Proceeding of the State Soil Science Conference WA 2019’, Crawley, WA. https://www.soilscienceaustralia.org.au/wp-content/uploads/2019/11/Programme-WA-Soils-conference-4-6-Dec-2019.pdf

Azam G, Gazey C, Bowles R, D'Antuono M (2019b) Recurring lime applications to fix acidity in the whole soil profile. Grain Industry Association of Western Australia. 2019 Perth Research Updates. http://www.giwa.org.au/2019researchupdates

Azam G, Gazey C, Bowles R, D'Antuono M (2019c) Combined application of lime and gypsum boosts grain yield in acidic soil. Grain Industry Association of Western Australia. 2019 Perth Research Updates. http://www.giwa.org.au/2019researchupdates

Azam G, Gazey C (2019) Re-engineering soil pH profiles to boost water use efficiency by wheat. In ‘Proceedings of the Agronomy Conference 2019’, Wagga Wagga, NSW. Available at http://agronomyaustraliaproceedings.org/images/sampledata/2019/2019ASA_Azam_Gaus_351.pdf

Azam G, Gazey C (2020) Slow movement of alkali from surface-applied lime warrants the introduction of strategic tillage for rapid amelioration of subsurface acidity in south-western Australia. Soil Research 59, 97-106.

Davies SL, Armstrong R, Macdonald L, Condon J, Petersen E (2019) Soil constraints: a role for strategic deep tillage. In ‘Australian agriculture in 2020: from conservation to automation’ (Eds J Pratley, J Kirkegaard) pp. 117–135 (Agronomy Australia and Charles Sturt University: Wagga Wagga, NSW). Available at https://www.agronomyaustraliaproceedings.org/index.php/special-publications

Davies S, Hagan J, Reynolds C, Smee V, Lefroy W, Andrew J, Newman P (2015) Long-term responses to soil amelioration – benefits will last more than 10 years! 2015 Agribusiness Crop Updates, 24-25 February, Perth, WA. Department of Agriculture and Food and Grains Research and Development Corporation.

Gazey C, Andrew J, Griffin E (2013) Soil acidity. In ‘Report card on sustainable natural resource use in agriculture,’ Department of Agriculture and Food, Western Australia. Available at https://www.agric.wa.gov.au

Kirkegaard J, Kirkby C, Oates A, van der Rijt V, Poile G, Conyers M (2020) Strategic tillage of a long-term, no-till soil has little impact on soil characteristics or crop growth over five years. Crop and Pasture Science 71, 945-958.

Li GD, Conyers MK, Heylar KR, Lisle CJ, Poile GJ, Cullis BR (2019) Long-term surface application of lime ameliorates subsurface soil acidity in the mixed farming zone of south-eastern Australia. Geoderma 338, 236–246.

Reynolds C, Azam G, Davies S, Gazey C, Parker W, Walker J (2021) Liming – It’s a no brainer. Grain Industry Association of Western Australia. 2021 Perth Research Updates. (accepted)

van Gool D (2016) Identifying soil constraints that limit wheat yield in South-West Western Australia. Resource management technical report 399. Department of Agriculture and Food, Perth, Western Australia. www.agric.wa.gov.au

Acknowledgments

We are thankful to the Government of Western Australia and the Grains Research and Development Corporation for investing in this work. Many thanks to all growers who have hosted our experiment in different parts of WA. We thank Gavin Sarr, Daron Malinowski, Richard Bowles, Sultan Mia, Alistair Hall, Kanch Wickramarachchi, Ross Gazey and DPIRD Field Research Service Team for assisting in technical aspects of the experiments. We also thank Dr Andrew van Burgel for helping with the statistical analyses.

Contact details

Dr Gaus Azam
Department of Primary Industries and Regional Development
75 York Rd, Northam WA 6401
Ph: (08) 8 9690 2159
Email: gaus.azam@dpird.wa.gov.au