Grains Research and Development

Developing and assessing agronomic strategies for water repellent soils

Stephen Davies, Paul Blackwell, Derk Bakker, Craig Scanlan WA Department of Agriculture and Food; Margaret Roper, Phil Ward CSIRO Plant Industry

Key messages:

  • Short-term soil water repellence mitigation options, such as improved furrow sowing and use of banded soil wetting agents, offer a relatively low cost way of improving crop establishment and performance across all the water repellent soils in a seeding program.
  • Long-term soil water repellence amelioration options, such as clay spreading or delving, rotary spading and mouldboard ploughing, are more costly and generally only applied to relatively small parts of the cropping program each year. They potentially overcome the water repellence for 5-10 years or more and have associated agronomic benefits and risks.

Aims:

  • To evaluate the efficacy, cost, value and other associated benefits and risks of the range of soil water repellence management strategies in broadacre cropping systems.
  • To give growers more confidence to choose which management option or combination of options best suits their needs, soil types, landscapes and circumstances.

Background:

Increased industry concern about water repellence may well be mainly due to accumulation of hydrophobic waxes in relatively undisturbed topsoils and more frequent dry autumns. Numerous options are available for managing soil water repellence for crop production. Mitigation options are usually low-cost and assist crop establishment in the short-term but need to be repeated each year. Amelioration strategies, which displace or overcome the soil water repellence, have a high-cost are slow to implement, but the benefits may last for up to 5-10 years or more (Table 1).

Table 1. Principal water repellence management options. (* indicates need for further evaluation)

Type Management Option Approximate cost Longevity Mechanism
Mitigation



Improved furrow sowing Cost of winged points or boots, press wheels Short term, months Grading of repellent soil into ridges and water harvesting
Banded wetting agent $10-12/ha/year; perhaps new presswheels* Short term, months* Aids water penetration into furrow base
Blanket wetting agent +/- water adsorber $25-50/ha/year depending on rate Short term, 1-2 years* Aids water penetration into and retention in topsoil
Full stubble retention, low disturbance seeding Possibly disc openers and more precise autosteer Ongoing* Water entry via remnant root pathways
Amelioration


Rotary spading $150/ha 3-7 years* Soil heterogeneity provides pathways for water entry
Soil inversion $100-120/ha Up to 10 years or more* Inversion of wettable subsoil layer to the surface
Clay spreading or Clay delving $300-900/ha 10-15 years or more Higher soil surface area & clay content masks repellence

Combinations of these strategies may be appropriate according to the relative area of repellent soil and the long term financial planning of the enterprise. More productive soils may be progressively ameliorated first; those with more weed control problems and less rocky subsoil being carefully inversion ploughed. For the balance of the cropping program the most appropriate mitigation methods can be employed. Relatively small problem areas of weed invasion and gravel outcrop may be more appropriately treated with blanket applied products or claying. Some mitigation strategies such as zero-till full retention systems involve a change of cropping system allowing improved infiltration and crop establishment and may be effective in the long-term if the system is maintained (Roper and Ward, 2011). Severe water repellence on soils with inherently low yield potential may require alternative land use strategies such as use of perennial pastures, fodder vegetation or trees. Determining the cost and benefits of options that involve changes in the farming system is complex and involve associated agronomic benefits, risks and impacts on profitability and the environment; economic modelling is being developed for this.

Method:

We aim to draw on the historic and current research to review our current understanding of most of the major management options available and to indicate of where and how they may best fit within a cropping system. Six research trials have been established over the last 3 years. Together these trials assess the full range of major management options (see Table 1) across a number of environments and seasons. In addition to these, many on-farm demonstration trials have been monitored and assessed with assistance of collaborating growers, grower groups and agribusinesses. In this overview we are drawing on the major findings and outcomes of all of these trials and industry experience. The 2011 season with summer rain and some substantial rainfall events were not conducive to expression of water repellence in many locations; however water repellence impacts could still be seen on more severely repellent soil types.

Results:

Improved furrow sowing

When furrow sowing was first developed for water repellent soils in the early 1990’s most of the seeding was being done with sweep or winged type points which tended to grade the water repellent soil out of the furrow into the ridges. There has been widespread adoption of knife points for seeding since the mid 1990’s. Recent observations have noticed that in many cases the furrows in water repellent sands were not wetting up and remaining dry, resulting in patchy crop establishment. Subsequent assessment has confirmed this finding with soil moisture measures often showing that in knife point systems the furrow is drier than the ridge (Table 2).

Table 2. On-farm comparison of a knife point seeder compared with a winged point, paired row seeder on water repellent sand and sandy gravel at Badgingarra in 2011. Yield differences were determined from paired hand harvest index cuts in discrete water repellent patches that were too small for meaningful replicate samples to be collected hence no statistical analysis is possible.

Soil type

Molarity ethanol droplet

Seeder type

Volumetric soil moisture (%, 0-10cm)

Crop establishment (plants/m2) Number of heads/m2 Grain yield (t/ha)
Furrow Ridge
Pale deep sand

1.5 moderate repellence Knife points - - 97 180 1.19
Winged points + paired rows - - 201 298 1.79
Sandy gravel 4.7 severe repellence Knife points 2.2 4.8 95 299 3.98
Winged points + paired rows 4.2 2.6 222 387 5.42

It is hypothesised that this problem may be due to dry water repellent soil falling behind the knife point during seeding and into the slot with the seed and fertiliser ahead of the closer plate. This problem would be exacerbated by dry sowing and fewer and smaller rainfall events at the break of the season. Further field research will be conducted in 2012 to better understand the principles and alterations that can be made to improve furrow sowing in water repellent soils. Increasing the flow of water repellent soil out of the furrow through use of winged points or seeding boots, press wheel shapes that encourage stable furrows and use of banded wetting agents are key strategies that may help.

In an on-farm seeder comparison, in furrow moisture, crop establishment, head number and grain yield were all substantially improved when winged points and paired rows were used compared with a knife point seeder (Table 2). It should be noted that the yield responses measured with paired cuts on water repellent areas were quite large compared to the average grain yield advantage for the winged point-paired row system across the whole paddock (200 kg/ha).This suggests that most of the overall yield gain may have been obtained from the increases in the relatively small water repellent areas. While it was not possible to determine which aspect of seeder design was most important from a seeder comparison it does demonstrate the type of benefits that an improved seeding system can have. In a research trial in 2011 at Balla, north of Geraldton, that included a comparison of knife points only with wings on the same knife points there was a trend to higher crop establishment and more tillers with the use of winged points but no difference in grain yield (Table 3). The advantage of improved furrow sowing is that it can potentially be applied over all the repellent cropping areas at relatively low cost, if only points and press wheels need changing.

Table 3. Impact of seeding point type and banded wetting agent on crop establishment and yield on low water repellence surface soil in a relatively wet growing season (252mm)

Seeding point type

Banded wetting agent

Water droplet penetration time (secs)

Crop establishment (plants/m2) Shoot number/m2 in June 2011 Grain yield (t/ha)
Knife points No banded wetting agent 15 172 223 2.55
Winged points 8 191 273 2.56
Knife points With banded wetting agent 2 209 285 2.53
Winged points 2 188 268 2.50
l.s.d. (0.10)   37 62 0.35

Banded soil wetting agents

Banding wetting agents in the base of the furrow, behind the press wheels, to assist water entry into the seed zone is an efficient use of surfactants as relatively low volumes of 1-2 L/ha are required, thereby reducing the cost. The efficacy of banded wetting agents can be variable with lack of success is usually a result of furrow infill or soil movement disrupting the application and preventing a continuous band of surfactant being applied to a stable soil surface. Press wheel shape is important, with V-shaped or rounded press wheels creating more stable furrows than rectangular section wheels. Banded wetting agent with knife points and press wheels for patchy dry sowing at Balla in 2011 increased wheat plant numbers by 37 plants/m2 with a corresponding increase in tillers of 62/m2 (Table 3). While this didn’t translate to a yield difference it does demonstrate the opportunity to improve crop establishment with this tool.

Blanket-applied soil wetting agents

In the early 1990’s researchers trialled the use of blanket-applied wetting agents at rates of 50 L/ha in broadacre cropping and pasture systems. Results were variable but improvements in crop establishment and yield were seen at some sites. However the cost of these treatments at the high volumes required and the risk of increased nutrient leaching and poor water retention from these longer-lived products also resulted in some negative responses. In a trial at Balla in 2011, a leaching year, application of blanket wetting agent to a yellow sandy earth gave no yield advantage, with grain yield tending to be reduced by 275 kg/ha (data not shown). Recent blanket applied wetting agent formulations have included humectants that retain moisture thereby reducing water retention problems. They persist longer giving about two years of benefit, improving economic return. Recent research indicates, however, that some of these formulations only work on quite specific soil types such as the forest loamy gravels or on firmer soils with reasonable clay content and are less effective on the repellent sands. Some products need reasonable rainfall after application to help them enter the soil and work effectively. Growers need to be cautious of the cost given the higher volumes required. Formulations containing water holding compounds may be an advantage as they reduce problems associated with reduced water retention that can occur with persistent surfactants. Further research and product development may expand the opportunities for this approach and it may have a role in improving establishment in smaller water repellent patches, such as gravel hill tops and old sheep camps, especially those with high weed burdens.

Full stubble retention, low disturbance and on-row seeding

Overcoming water repellence with minimal soil disturbance and full stubble retention systems appears to involve the maintenance of residual root systems and preferential flow paths that act as conduits for water entry into water repellent soils. This is despite the fact that retention of stubbles can lead to increases in soil organic matter in the surface soil and consequently increased soil water repellence. In a 4-year study, water infiltration and water repellence, soil C and crop performance were monitored in treatments comparing zero-tillage versus annual cultivation of the topsoil and stubble retention versus stubble removal by burning or grazing. Soil water repellence, as measured by the Molarity of Ethanol Droplet (MED) method, was higher in the stubble retention and zero-tillage treatments (Figure 1A) that had higher soil C levels than cultivation and stubble removal treatments (data not shown). The worst repellence occurred under zero-tillage and stubble retention and least under stubble removal and cultivation (Figure 1A; R2 for %C versus MED ranged from 0.75-0.83). However, soil water content contradicted the findings on water repellence and indicated that water infiltration was best under zero-tillage and stubble retention and poorest under stubble removal and cultivation (Figure 1B), and this impacted on crop performance.

Graph A measures water repellence against stubble treatment at depths of 0 to 5cm and 5 to 10cm, with the lowest water repellence across both depths measured with burnt stubble and cultivation, at 1 MED at 0 to 5cm and 0.7 at 5 to 10cm. Repellence at 0 to 5cm decreases evenly from Retained NT, Retained CT, Burnt NT, Burnt CT, but 5 to 10cm depth repellence is more variable across treatments. Graph B demonstrates th ehighest soil water content% (v/v) with NT stubble retained at 9.7%, with progressively less across CT retained, and NT burnt and CT burnt both very close to 7.4%. 

Figure 1: Soil water repellence (A) and soil water content (B, 0-10 cm) in a sandy water repellent soil under stubble retention and zero-tillage (Retained NT), stubble retention and cultivation (Retained CT), stubble burnt and
zero-tillage (Burnt NT), stubble burnt and cultivation (Burnt CT).

Clearly this approach does not necessarily remove or alter the properties of the water repellent soil but improves water entry into the soil. Visual evidence indicates that under zero-tillage, bio-pores and old roots are preserved and these provide preferred pathways for water movement into the soil.

Growers have also observed that they can get better crop establishment when they seed on or very near to the  previous season’s crop row. In a wetting agent trial at Balla in 2011, dry sown lupin establishment was improved through the use of banded wetter when seeded between the previous seasons wheat rows. However, much greater improvements in establishment were achieved by seeding on the previous season wheat rows (Fig. 2). Research is ongoing to examine this but maximising these benefits may involve growing crop species or having soil conditions that promote dense, extensive root systems. Establishing sufficient plants on repellent soils can be difficult in the first place so other tools may need to be used in conjunction with this system. An experiment which includes a rotary spading treatment that is sown with a zero-till disc seeder has been established with the aim of promoting extensive root development by spading then maintaining this with minimal disturbance. On-row seeding has potential but stubble handling and other issues may limit its application, especially when seeding with points.

On row establishment with or without a banded wetter achieves 36 plants per square metre, while between row establishment is less successful at 20 plants with a banded wetter or 6 plants without.
Figure 2. Dry sown lupin establishment when sown either between or on the previous year’s wheat rows and with or without (Nil) banded wetting agent at Balla on deep yellow water repellent sand in 2011.

Rotary spading and mouldboard ploughing

The uptake of rotary spading and mouldboard ploughing (soil inversion) for one-off weed control and soil renovation has continued to increase. In 2011 over 10000 ha of sandplain soil was inverted using the mouldboard plough primarily for the control of herbicide resistant weeds. Wind erosion, seed being sown too deep, recompaction and sandblasting remain the biggest risks with these tools. Wind erosion can only be minimised by spading or ploughing the soils when they are wet and immediately sowing a cereal cover crop. Sands on the south coast which are very fine with even grain size distribution have proven to be particularly vulnerable to wind erosion. Many growers who are mouldboard ploughing are using simple light seeder bars and sowing shallow or broadcast spreading the seed and pressing it in with coil packers or press wheels.

Controlled traffic will be very important for maintaining long term benefits of these effects. Grain yield responses in the first year or two after spading or mouldboard ploughing have generally been good. The average wheat grain yield response to mouldboard ploughing for 16 comparisons in the first year is 544 kg/ha while for 3 years after mouldboard ploughing it is 338 kg/ha. For spading the average wheat grain yield increase is 603 kg/ha for 12 comparisons in the first year. Additional significant findings from 2011 are as follows;

  • One-off inversion ploughing and spading results in more roots and higher soil moisture levels at 15–30 cm but cultivation dries the topsoil in the short-term. Burial of topsoil nutrients and organic carbon indicate a need for soil testing to 30 or 40 cm before and after amelioration (See Scanlan et. al. 2012, these proceedings).
  • The negative consequences of poor soil inversion with a mouldboard plough resulting in concentration of repellent soil in the seed zone could be partially overcome with the use of a banded wetting agent. This improved grain yield by 270 kg/ha over mouldboard ploughing without the use of wetting agent in a trial at Balla in 2011. It should also be noted that this occurred despite above average rainfall at the site, 252mm growing season and177mm out of season.
  • Inversion ploughing significantly reduced the incidence of leaf disease (septoria nodorum blotch and yellow spot) in wheat in the first year compared with an untreated control.

Clay spreading

Clay-spreading involves the spreading of clay-rich subsoil onto the soil and this subsoil is then incorporated into the water repellent topsoil. Clay spreading on water repellent soils was first done by South Australian growers in the late 1960’s. These growers and numerous researchers since have found that using moderate application rates of 75-150 t/ha of dispersive clay-rich subsoil (25% or more clay) can be easily incorporated and is effective at overcoming soil water repellence. Recent research has shown that a topsoil clay content of 3% is sufficient to overcome topsoil water repellence where the organic carbon content is 1% but higher clay contents of 5-7% are required where the organic carbon content is higher at 1.5% (DAFWA Bulletin 4773, 2009).

Table 5. Summary of wheat yield responses to spreading of moderate rates of clay–rich subsoil in claying trials and demonstration sites established over the last 3 years. Grain yield responses that were not statistically significant are indicated by ‘ns’.

Location Clay-rich subsoil application rate (t/ha) Clay % In subsoil Incorporation method Year Years after claying Wheat Yield response (kg/ha)
Bolgart

260 38 offset discs 2010 1 250
2011 2 560
Badgingarra

150 31 offset discs 2009 1 270
2011 3 460
Badgingarra (demo)

100 31 offset discs 2009 1 420
2011 3 536
Badgingarra (demo) 100 - offset discs 2011 1 358
Binnu

125 26 spader (shallow) 2009 1 200 ns
2011 3 439
Balla 150 52 spader (shallow) 2011 1 21 ns

Despite claying being a very effective option for overcoming non-wetting soils the high cost and potential risks associated with the practice have limited its adoption. High clay rates at the surface can result in surface sealing or water being held in the topsoil where it is subject to evaporation and can reduce root development into the subsoil. Farmers and researchers have often found that crops on clayed paddocks grow large biomass but are prone to haying off and yielding poorly, particularly with poor incorporation and in drier seasons or environments. Such growth can, however, provide useful livestock feed. Compaction from the passage of heavy subsoil-laden machinery during spreading and use of subsoil with high levels of boron, salt or can also result in yield reductions due to claying. Recently established trials have shown that claying often increases wheat yield but that responses tend to be less in the warmer and shorter season environments in the northernmost part of the wheatbelt (i.e. Binnu and Balla) where holding moisture near the surface may result in higher evaporative losses (Table 5). Inputs of potassium and sulphur from the applied subsoil can be significant and subsoils should be tested. The cultivation involved in incorporating the clay-rich subsoil into water repellents topsoils can be a benefit in its own right. The spreading of moderate rates of clay-rich subsoil that is easily incorporated and free from toxicities (e.g. high pH for lupins) in higher rainfall environments continues to provide a good tool for long-term amelioration of water repellent sandplain soil with additional advantages of improved nutrient holding and reduced erosion risk.

Conclusion:

The current soil types and locations suitable for each of the management methods are summarised as follows: Improved furrow sowing is suitable for all locations and appears to be suitable for water repellent deep sands, sandy duplex and sandy gravel soils but needs testing on the forest loamy gravels. Banded soil wetting agents may be suitable for all locations and all soil types. It provides some insurance that crop establishment will be optimised and may be particularly beneficial for dry sown crops on water repellent soils in certain circumstances. Current blanket soil wetting agent products tend to be suited to forest loamy gravels and firmer soils with reasonable clay content. Historical trials indicate that blanket application of predominantly surfactant-based wetting agents typically reduced repellence but sometimes reduced yield due to enhanced leaching and reduced water retention. Full stubble retention and on-row seeding should work on all soil types but may be better suited to higher rainfall, longer season environments where crop biomass, crop density and extensive root development can be achieved to maximise the effect. One-off soil renovation with rotary spading and mouldboard ploughing has worked well on deep sandy earths, pale deep sands and sandy gravels. The tools should work on deep sandy duplex but not shallow duplex soils and cannot be used in soils with abundant rock or cemented gravel. In very weak pale sands these tools can improve crop establishment and biomass but poor water holding capacity can still limit productivity and it appears the fine sands on the South coast may be too sensitive to wind erosion for these tools. Clay spreading should work on most soils but may not be suited to warm, shorter season environments as early water use and increased evaporation may limit water availability at grain filling. It still has benefits for livestock feed and ground cover.

Given the nature of the range of options available the best approach may be for growers to utilise improved furrow sowing and or banded surfactants as mitigation tools that can work across a wide range of soil types and over the whole seeding program for relatively low cost. The higher cost and longer lasting amelioration options such as clay spreading or one-off soil renovation with rotary spading or soil inversion could then be applied to smaller areas of strong water repellence or areas where the biggest productivity gains are likely to occur.

Keywords:

Water repellence, furrow sowing, soil wetting agents, soil inversion, rotary spading, claying

Acknowledgements:

GRDC and DAFWA fund "Delivering agronomic strategies for water repellent soils in WA" and GRDC and CSIRO fund the "Novel solutions for managing non-wetting soils" projects. Collaboration of Bill Bowden, David Hall (DAFWA) and the West Midlands, Liebe, Northern Agri Group, Mingenew-Irwin, WANTFA, SouthernDIRT and Stirlings-to-Coast grower group staff and growers is acknowledged. DAFWA technical support - Breanne Best, Grey Poulish, Trevor Bell, Dirranie Kirby and Steve Cosh; Ciara Beard and Anne Smith (DAFWA Geraldton) for assessment of leaf disease. CSIRO technical support: Angela Muller, Shayne Micin, Nicola Palmer, Julianne Hill (DAFWA), Jenny Chambers (RAIN).

GRDC Project No.:   DAW00204 and CSP139

Paper reviewed by:  Martin Harries and Bob French