Module 01 >

Monitoring Soils

Module 02 >

Interpreting Soil Results

Module 03 >

Soil Fertility and Nutrient Management

Module 04 >

Common Soil Constraints

Module 05 >

Soil Carbon Capture

Module 06 >

Digital Agriculture for Soils

Module 07 >

Using Biologicals to Build Soil Organic Matter and Resilient Soils

Module 08 >

Managing Irrigated Soils in the Riverina Region of NSW

04

Common Soil Constraints

Module 01 >

Monitoring Soils

Module 02 >

Interpreting Soil Results

Module 03 >

Soil Fertility and Nutrient Management

Module 04 >

Common Soil Constraints

Module 05 >

Soil Carbon Capture

Module 06 >

Digital Agriculture for Soils

Module 07 >

Using Biologicals to Build Soil Organic Matter and Resilient Soils

Module 08 >

Managing Irrigated Soils in the Riverina Region of NSW

SECTION
4 A

icon_section_3a-

What are soil constraints?

What are management options for soil constraints?

Soil constraints usually fall into one of three camps.
1. Mitigation - short term fixes (often cheaper)
2. Amelioration - fixing the problem for good
(more expensive)
3. Land use change

Soil constraints are a range of limiting factors present in the soil environment that can adversely impact plant growth, agricultural productivity, and ecosystem health.

These constraints could be physical, chemical, and biological factors that collectively affect soil structure, nutrient availability, water retention, and overall soil fertility.

Soil constraints have an annual yield penalty costing billions in lost income. In wheat alone, soil acidity is estimated to cost A$400 million each year (Orton et al. 2018). Efforts into ameliorating soil constraints has had considerable focus over the last fifteen or so years, but there is still much to learn.

Understanding and managing soil constraints are vital for sustainable land use and effective crop management, as they directly influence the success of agricultural operations. land restoration projects, and environmental conservation efforts.

By identifying and addressing these constraints, farmers, land managers, and researchers can optimise soil conditions and promote healthy plant growth, leading to improved productivity, and resilience in various landscapes.

Let’s look at some of the key soil constraint factors that can limit soil functionality, their impact, and potential mitigation strategies.

Every farm is different, and every paddock will need its own prescription for managing the constraints in that paddock. In some cases, the problem is so severe that areas of the farm would be better put to a different use or at least taken out of production. The cost of amelioration far outweighs the returns from land improvement. In very poor-performing areas, even just stopping production saves on input costs. Highly saline, severely waterlogged areas and those affected by micronutrient toxicities are examples. In this module, each soil constraint is addressed individually, however in real life, like in the case study , there could be complex situations with multiple constraints.

SECTION
4 B

Finding constraints on the farm

The first step to ameliorating soil constraints is knowing where they are.

As grid soil sampling is cost-prohibitive in Australia, but many growers and advisers use spatial data such as yield maps to divide paddocks into zones and find poorly performing areas that could be due to soil constraints. More information on how to use spatial data is in Module 7: Precision Agriculture for Soils.

After years of seeing a farm through wet and dry seasons, most growers have a very good idea of where the problem areas on their farms are. This information combined with spatial data is a good way to narrow down the search area for constraints.

Remember

The first step to ameliorating soil constraints is knowing where they are.

Tackling amelioration on variable soil types outlines one way to map constraints on the farm and choose which areas to ameliorate. The Dispersive Soil Manual outlines a process specific to dispersive soils and their associated constraints. Both methods involve using aerial maps, yield maps, soil maps etc. to find problem areas, then ground truthing with soil tests to confirm the extend and severity of constraints. This first step can be laborious but necessary to avoid wasting time and money ‘fixing’ a problem that isn’t there.

SECTION
4 C

Soil acidity

What is an ‘acidic soil’?

Most crops have a preferred soil pH range. Outside of this range, crop growth slows because,

1. Nutrient availability is affected

In acidic soils, many key plant nutrients including nitrogen, phosphorus, sulfur, calcium, and molybdenum are less available for plant uptake.

2. Aluminium and manganese toxicity harm crop roots

If pH drops enough, manganese (<4.8 pHCaCl2) and aluminium (<4.5 pHCaCl2), become toxic to plant roots. High levels of aluminium harm the root; root tips are brittle, and root growth and branching are reduced (Gazey and Davies, 2009). Aluminium levels above 2 mg/kg start becoming a problem. Aluminium toxicity is usually below 10 cm depth as organic matter binds aluminium, limiting the impacts on crop roots.

Manganese toxicity is more common with lucerne, cabbage, cauliflower, cereals, clover, pineapple, potato, and tomato (Incitec Pivot Fertilisers, 2022). As with most trace elements, tissue tests are better to diagnose manganese toxicity than soil tests.

3. Microbial activity is reduced

With lower microbial activity:

Organic matter breakdown and nutrient release (mineralisation) slows.

Legume nodulation can fail. Most rhizobia bacteria prefer a soil pH > 5.5 but the individual species may vary in their tolerance to acidity. For example, medic rhizobia (Sinorhizobium spp.) prefer a pH closer to 7 while rhizobia on lupins (Bradyrhizobium spp.) can tolerate more acidic conditions.

fig-1-damage-to-roots

Figure 1. Damage to the root system, growth, and development of wheat (right) caused by aluminium toxicity in acidic soil, compared to the unaffected (left). (Source: OMEX Agriculture Inc. Canada.)

How to diagnose an acidic soil?

Measuring soil pH is the only way to diagnose acidity. This is best done with laboratory tests; however, the field kits can give a guide to soil pH and help you decide which areas need further investigation. As a minimum, test pH at 0-10cm, 10-20cm, and 20-30cm depths.

An acidic subsurface layer, for example from 2-4 cm, limits root growth in some soils. It can go undiagnosed because the 0-10 cm samples average out the soil pH. To check for subsurface acidity either break up the sample into 0-5 and 5-10cm, or if there is a soil corer, do an indicator test down the profile to help identify the acidic band.

Use crop guidelines to decide if soil pH is too low as different crops can tolerate different pH levels. Generally, a cropping soil is considered acidic if the pH is <5.5 from 0-10 cm and <5.0 below 10 cm.

How to manage an acidic soil?

Liming

Liming is the most cost-effective way to treat acidic soil. Without liming, the soil will continue to acidify, further penalising yields and costing more to fix.

Lime sources

Agricultural limestone, lime sand, and dolomite are common lime sources. Not all lime sources are created equal. How well a lime works depends on its particle size and neutralising value (NV). The higher the NV and smaller the particle size, the faster it works. Because lime is mined, the characteristics will vary between sources and even within pits. Comparing lime sources has more guidance.

There are multiple calculators (e.g., Southern Farming Systems Lime Assist calculator, DPIRD’s iLime) to help compare lime sources. These calculators include the costs of lime and transport, and ability to treat acidity, to help you decide on how much lime to add and which lime source is the best value.

Treating subsoil acidity

Top-dressed lime can take 10 years or more to start treating acidity below 10cm depth. If subsoil acidity is a problem now, the lime needs to be incorporated into the acidic layer(s).

Cultivar choice

Choosing cultivars more tolerant of acidity or alkalinity help maintain productivity.

fig-2-relaive-tolerance

Figure 2. Relative tolerance of crops and pastures to soil acidity and aluminium toxicity. (Source: DPIRD, 2018.)

SECTION
4 D

Soil salinity

What is soil salinity?

Saline soils have higher levels of salts in the soil water, making it harder for crops to extract water and nutrients. The higher the salinity the more energy plants need to take up water. If soil salinity is high enough (when the concentration in the soil is higher than inside the plant roots), water moves out of the plant roots and into the soil water, drying out the plant.

Saline soils can also cause chloride toxicity in plants. Germinating plants and seedlings are more susceptible to salinity than older plants.

There are multiple types and causes of salinity, which affect how they are managed.

Dryland salinity

Dryland salinity is the most common form of salinity in broadacre agriculture. It occurs when salts in the groundwater or already in the soil rise to the surface.

The issue is greater in clay soils more than sands, when evaporation is high (hot summers, bare soil) and when the water table is within 3m of the surface. Because clay soils have ‘tighter’ pores, they have greater capillary rise. Water and dissolved salts can move further up towards the surface than in sands with larger pores. As the water evaporates from the soil, the salts are left behind and concentrate over time, eventually becoming too toxic for crop growth. Bare areas area then prone to erosion.

From irrigation water

Irrigating crops with salty water will cause salts to build up in the soil.

Creek-line/Riverbank salinity

The soil along saline creeks and riverbanks tends to be saline. As it rains and the creeks swell, they can leave salts in the areas around the banks.

Temporary salinity

High gypsum, fertiliser and manure applications can temporarily inflate soil salinity. As long as high application rates do not continue regularly, these will drop back down again.

How to identify saline soils?

Testing the electrical conductivity (EC) of the soil is the best way to diagnose soil salinity. This is described more in . Generally, a soil with an EC reading <2dS/m is not saline.

There are some visual clues to salinity; however, these must be confirmed with soil testing.

  • Yellowing plants and burnt leaf tips
  • Tree dieback
  • Bare patches
  • Visible salt crusts
  • Stock licking the ground surface salt.
  • Waterlogged patches

EM38 surveys measure conductivity and can be useful to identify potential areas of salinity on the farm. Again, these need to be confirmed with lab tests to identify the extent and severity of salinity. EM38 surveys are discussed more in Module 7: Precision Agriculture for Soils

fig-3-Salt-crusting

Figure 3. Salt-crusting and soil erosion due to salinity. (Source: NSW Department of Planning and Environment)

Managing saline soils

The key to managing dry saline land is to stop salty water rising to the surface. This could be done by keeping the soil covered with crops and mulches to limit evaporation.

Strategies include:

Plant salt tolerant cultivar

Salt-tolerant crops are one of the best management techniques for saline soils. They will limit evaporation and salt rise.

Manage stock

Do not let stock over graze/bare-out areas.

Mulch

For small bare areas, mulch will limit evaporation from the soil. This is not economical on large scale.

Fix other soil issues

While there is no fertiliser or product to fix soil salinity, fixing other soil issues such as poor drainage and maintaining balanced crop nutrition
will give crops the best chance of survival in saline soils.

If the soil is both saline and dispersive ( see dispersive soils ), gypsum can help improve soil structure and facilitate leaching.

Manage surface water

Install contour banks or swales to manage surface water flows, and limit erosion and waterlogging.

Reduce waterlogging

If waterlogging is making salinity worse ( see waterlogging ), consider ways to reduce waterlogging such as surface water management, and drainage.

Leaching

In irrigated crops, leaching the soil by irrigating with non-saline water can wash excess salts beneath the root zone.

Take care – if the salinity has not come from saline irrigation water, over irrigating can lead to waterlogging and exacerbate the problem.

Land use change

If areas of salinity are severe enough, it might be worthwhile to remove the areas from production and revegetate the land.

SECTION
4 E

Dispersive soils

What are dispersive soils?

Dispersive soils are sometimes called Sunday country (too wet Saturday, too hard and dry Monday). This is because when dispersive soils are wet, the aggregates collapse, and the soil becomes boggy. It then dries into a hardened, structureless mass.

Dispersive topsoils develop a crust that inhibits seedling emergence and infiltration, causing run-off and erosion. Dispersive subsoils can be too dense for crop roots to grow through, limiting access to water and nutrients, and cause waterlogging as well.

Dispersive soils often have other chemical issues including salinity, high pH (alkalinity), boron and chloride toxicity, and nitrogen losses from denitrification.

Sodicity vs dispersion

The term dispersive soil is often used interchangeably with sodic soil, but the two are not synonymous. Sodic soil has exchangeable sodium levels >6%. Although too much sodium is the common driver of dispersion, soils do not always disperse once ESP exceeds 6%. There are soils that disperse with an ESP of 4, and those with an ESP of 18 that do not disperse. This is because other soil properties, including salinity, clay content, clay mineralogy, and organic matter content affect dispersion.

fig-4-sodic

Figure 4. A dispersing sodic soil in a waterlogged landscape, causing severe land degradation. (Source: ©2008 Timothy Overheu, DPIRD)

How to diagnose a dispersive soil?

Field clues

Some common signs of dispersive soil include:

In plants:

Poor or patchy establishment

Bare areas

Shallow roots

Lower yield than expected

In the soil:

Surface crusts

Cloudy puddles

Ground stays boggy

Hard subsoil

Slow water infiltration

fig-5-dispersive

Figure 5. Common signs of a dispersive soil (Source: DPIRD)

The best way to diagnose dispersion is to do a dispersion test by either ASWAT or EAT test. This involve placing aggregates in a shallow dish of distilled or rainwater, and watching if they disperse and/or slake. For instructions on how to do these tests, see Appendix B page 56 of the GRDC Dispersive Soil Manual.

fig 6 dispersion

Figure 6. Soil dispersion test

Management options for dispersive soils

Gypsum

Gypsum is the cheapest way and fastest way to treat dispersive soil. Gypsum works in two ways. In the short term it suppresses dispersion through salinity. This is known as the ‘electrolyte effect’. This effect only lasts a couple of seasons and is achieved with lower rates of gypsum, around 2-3 t/ha. For a long-term fix, ongoing gypsum applications eventually replace the sodium in the soil with calcium, allowing the soil to form more table aggregates.

Elemental sulphur

As dispersive soils are often alkaline, elemental sulphur (ES) will lower pH, but it is an expensive product that warrants careful consideration before use. On calcareous soils, lots of ES is needed to lower soil pH which can become cost prohibitive. Applying too much ES can make the soil acidic. Use is more likely in horticultural crops with smaller land areas and higher value production.

Lime

As a source of calcium, lime can help treat dispersion where the topsoil has an acidic or neutral pH. This is less common than alkaline dispersive soil. Lime won’t help if the soil is alkaline.

Crop choices

For subsoil dispersion, crops with strong root systems and a sturdy tap root can create root channels for subsequent crops. This is called ‘root priming’. Lucerne (Medicago sativa), sulla (Hedysarum coronarium) and chicory (Chicorium intybus) are often suggested as root-priming crops.

Organic matter

Research is underway on using organic matter to treat dispersive soil. Organic matter improves aggregate stability and as it decomposes, bumps up soil salinity which suppresses dispersion.  Trials at Rand, NSW found that placing wheat stubble and fertiliser at depth improved root growth, soil aggregation and yield.

At the moment, using organic matter to treat dispersion is not a cost-effective option if the material has to be brought in from off-site.

Controlled Traffiic Farming

In Controlled traffic farming (CTF), vehicles use permanent wheel tracks reducing wheel traffic from 40-70% down to 10-15% on other parts of the paddock (Condon G & K, 2016). On top of limiting compaction and preserving soil structure after ameliorating the soil, CTF systems tend to have higher yields and lower fuel use.

Minimise cultivation

Minimise cultivation

Cultivating dispersive soil is a perilous activity. Not only can ripping can bring clods of dispersive soil to the surface, but working dispersive soil can induce dispersion in mechanically dispersive soils (those that disperse after they’ve been worked).

If you want to rip or try to alleviate the subsoil density, do a test strip first and see what happens after it rains. If it makes the problem worse, it’s best not to rip any more.

In some instances, ripping is a way to place ameliorants such as gypsum and organic matter deeper into the soil, closer to the problem.

SECTION
4 F

Soil compaction

What is soil compaction?

Machinery and stock foot traffic compact the soil, reducing pore spaces for air and water flow, and making it physically difficult for crop roots to grow.

Compacted soil:

  • becomes denser, making it harder for crop roots to grow.
  • lowers seedling emergence.
  • leads to smaller, less vigorous seedlings.
  • has fewer pores for oxygen exchange.
  • restricts crop access to deeper nutrients.
  • reduces water infiltration.
  • is more prone to waterlogging.

Plough pans and traffic pans are the main types of ag-induced compaction. Plough pans form when repeatedly tiling the soil at the same depth. Traffic pans are caused by soil compression from vehicle traffic.

Did you know?

Soil compaction caused by livestock is generally within the top 15cm. Subsurface compaction is from vehicles and can be as deep as 50cm. (Soil Quality 6: Compaction)
fig-7-plants-growing

Figure 7. Plants growing in (a) soil with good tilth and (b) soil with compaction. (Source: Vic Kulihin, SARE, University of Maryland)

How to diagnose soil compaction?

Digging a hole in soil often reveals where the compaction is. The soil becomes noticeably harder to dig once you hit the compacted layer. In a soil pit, you can sometimes see where the compacted layer starts, while the layer itself will look dense and blocky. 

Measuring soil compaction is best done with a penetrometer. Dry soil can appear compacted, so penetrometers are best used whenthe soil is wet (not saturated). Readings above 1.6MPa indicate compacted soil.

Instead of a penetrometer, metal rod or sharp stick will help you find the depth the hardpan starts.

Staggered crop root growth can be a clue to compaction. Soil compaction could be the issue if:

  • roots suddenly stop growing at a certain depth
  • root tips are stubby and swollen instead of more pointed

However, root issues could also be from other problems such as acidity or a nutrient toxicity. Test soil salinity and acidity just below where root growth stops to rule out these potential chemical issues.

What is soil compaction vs hardsetting?

Soil hardsetting is different to soil compaction. Hardsetting can occur naturally, even if traffc is perfectly managed. Soils most prone to hardsetting are either:

Dispersive clayey soils (see )
Sandy soils with low OM (<2%) and very weak aggregates

Very sandy soils are typically devoid of soil aggregates (or peds). When wet, the aggregates swell and collapse (slake) so when the soil dries it becomes a hardened mass with little pore space. Hardsetting sandy soils become signifcantly harder (and problematic for root growth) as the soil dries. Penetration resistance increases exponentially as the soil dries, the soil hardening to a degree that crop roots cannot grow any deeper. The soil softens again when wet.

Managing hardsetting soils is a challenge. Even if ripping alleviates compaction, if the soil is prone to hardsetting it could become a problem for root growth the next time the soil wets then dries out.

Managing soil compaction

The right management option depends on the cause of the compaction.

Mechanical intervention

Busting up the compacted layer is often the fastest way to help roots grow deeper but does mean tilling the soil.

Ripping is the most common way to deal with compaction in sandier soils. The heavier (more clayey) a soil gets, the more variable the response and ripping can even result in yield penalties. Though any tillage equipment can work as long as it can work into or even better below the compacted layer.

Once ameliorated, traffic management is essential (e.g., CTF) as up to 80% of soil re-compaction happens in the first pass of the machinery.

Implementing controlled traffic farming (CTF) is the best way to avoid compaction. In CTF vehicles use permanent wheel tracks, reducing wheel traffic across other parts of the paddock. Once on CTF, tramlines are like roads and need managing. Over time they will compact, become rutted and difficult to traffic, and will need renovating to keep the tracks at a similar elevation to the rest of the paddock. The CTF technical manual has further advice for broadacre cropping.

Crop choices

Crops with strong root systems and a sturdy tap root can create root channels for subsequent crops. Lucerne (Medicago sativa), sulla (Hedysarum coronarium) and chicory (Chicorium intybus) are often suggested as root priming crops. No-till and CTF will help maintain the channels once they have formed.

Avoid compaction

Avoiding compaction in the first place is ideal. Don’t drive on paddocks when they are wet if you can avoid it. Use optimal tyre pressure. Over-inflated tyres cause more soil compaction, and under inflation reduces tyre life.

SECTION
4 G

Non-wetting soils

What is a non-wetting soil?

In non-wetting or hydrophobic soil, water struggles to infiltrate. Without water, crops will not germinate. Where water does infiltrate, it follows old root channels which are less hydrophobic, leading to patchy crop emergence. This makes in-season crop management activities such as spraying and fertiliser harder because plants are at different growth stages. It can also mean a greater risk of erosion on the bare patches.

Water repellence is caused when organic materials such as plant waxes coat individual soil particles, making the repel water. Fungal hyphae may also cause soil water repellence, but their influence is not entirely certain and seems to vary with species (Davies et al. 2022)

Repellence usually develops on sandy soils (<5% clay). It is unlikely to develop on clayey soils because clays have a much higher surface area.

fig-8-convex-edged

Figure 8: Typical convex-edged water drop suspended on repellent soil (Source: DAFWA)

Identifying a non-wetting soil

Water droplet penetration test

The fastest way to test for repellence is to gently place a few drops of water on the soil surface. If the soil is repellent, the water will sit on top of the soil. The time it takes to filter in is a guide to how repellent the soil is. Water will appear to infiltrate but repellence is still lurking just below the surface.

table-moderate-severe-water

Molarity of Ethanol Droplet (MED) test

The MED test measures the concentration of an ethanol solution required to penetrate the soil in under 10 seconds. This test can only be done on dry soil as wetting the soil reduces repellence. It is also best done after scraping away the top 5 – 10mm of soil, because the sun can volatilise the waxes causing repellence at the very surface. In this video, Glenn McDonald from DPIRD shows how to do the MED test.

Paddock clues

In the paddock, clues to water repellence are:

  • Patchy or crop and weed emergence
  • Staggered crop growth
  • Herbicides not working well

Water repellence tends to only form in the top few centimetres of soil. It is more likely to develop in minimum or no-till systems.

Did you know?

In low rainfall areas, the water repellent layer acts like a mulch, reducing soil evaporation. Repellent soil can also channel water into furrows. These benefts are generally outweighed by the problems repellence causes.

Management options for non-wetting soils

Wetting agents

Wetting agents are a short-term option to help water infiltrate in the season they are applied. They are cheaper than long-term fixes (e.g., claying) and a useful ‘band-aid’ solution while working out an amelioration plan and budget for the farm.

On-row sowing

Water infiltrates more easily into old crop rows. Sowing on or within a few cm of previous crops rows puts seeds in more wettable soil, giving it a better change of establishing. This method only works in no-till systems where old crop rows are still present. GPS guided tractors are necessary.

Furrow sowing

Furrow sowing creates small, alternating ridges and furrows. Slopes of water-repellent ridge surfaces allows water to run-off and collect at the base furrow where seed are sown, improving water infiltration to soil. Use of winged points is more effective for furrow sowing as knife points can be less effective because repellent soil flows around the point, into the sown furrow.

Seeding options

Using higher seeding rates, cross-seeding (sowing at an angle to
last year’s crop) and delayed seeding in the hope of rain are options
to increase crop establishment in non-wetting soil.

Claying

Claying means spreading and incorporating clay-rich soil into the sandy topsoil. Claying permanently fixes water repellence, as well as increasing soil water holding capacity. It is expensive, costing hundreds of dollars per hectare – but if done right, fixes the problem.

Soil inversion, spading or delving

Like with claying, the goal with these practices is to increase the clay content of the soil and dilute the non-wetting soil.

  • Soil inversion flips the soil, burying the non-wetting soil and bringing wettable soil to the surface.
  • Spading mixes the non-wetting soil and wettable subsoil together, diluting the repellent soil.
  • Delving brings up seams of wettable subsoil to the surface.

For these methods to work, the subsoil mixed in or brought to the surface must have enough clay to increase the topsoil clay over 5%. If it doesn’t, water repellence can re-develop.

For more information on claying, inversion, or delving see Spread, Delve, Spade, Invert.

Liming

If soil acidity is also a problem, the pH boost from liming encourages actinobacteria which degrade the waxes causes repellence. This can reduce water repellence but is not an economical option if the soil is not acidic.

SECTION
4 H

Waterlogging

What is a non-wetting soil?

Waterlogging happens when drainage is poor, and the excess water fill in all the pore space in root zone. The lack of air inhibits gas exchange of roots and soil biological activity, inducing oxygen deficiency in roots and activity of anaerobic microbes.

Too much water in soil is a major constraint to crop growth and is sometimes caused by other soil constraints such as compaction and dispersion. Waterlogged soils can also stem from high water tables or active underground springs.

Waterlogging is more common on clay and shallow duplex soils. It can be tricky one to catch if the subsoil is waterlogged but you can’t see it on the surface.

Issues for soil:

  • poor trafficability
  • nutrient loss e.g., leaching, denitrification
  • soil structural decline

Issues for crops:

  • anaerobic conditions (not enough oxygen) causing:
    • a build-up of hydrogen sulphide which damages crop roots
    • root death from limited oxygen
  • induce or exacerbate manganese toxicity
  • increased salt uptake. Crops take up more salts in waterlogged soils. This is a bigger problem is saline soils and can lead to plant death.
  • harder to take up potassium and magnesium
  • slow growth because waterlogged soils stay colder
  • more weed competition
  • reduced fodder quality

How to diagnose waterlogging?

In the paddock:

  • water ponding on the soil surface
  • boggy patches
  • slipping and getting stuck when driving on the soil
  • yellowing crops and pastures
  • certain weeds such as docks, goldenrod, chickweed, sedges, and horsetail indicate poor drainage.

In the soil:

Dig a hole and look for:

  • grey or greenish coloured soil for long-term waterlogging
  • mottles for periodic waterlogging (cycles between wet and dry)
  • if water flows into the hole while you’re digging

Management options for waterlogged soils

Crop choice

In pastures, crop such as tall fescue and Phalaris are more tolerant of waterlogged soil. Consider cultivars more tolerant of waterlogging.

Manage grazing

Keep stock off waterlogged areas to avoid soil structural issues and further drainage problems.

Avoid the area

If the waterlogged area is small enough it might be worth removing it from production and avoiding it completely. This is more likely in  areas where a high water table or active spring are the cause, and mitigation options are limited.

Improve drainage

Surface or subsurface drains reduce waterlogging but come with substantial costs. Weigh up the cost of drainage against the frequency, area, and severity of waterlogging.

Good drainage design needs careful planning, considering:

  • The cause and frequency of waterlogging
  • Outfall – can water drain by gravity or will you need a pump?
  • Which areas to drain first
  • Soil type
  • Suitable drainage systems
  • The wider impacts of water redirection, including impacts on roads and neighbouring properties.

Surface drains are usually considered frst where they can re-direct surface water that is exacerbating waterlogging. Common options include trench drains, spoon drains, and hump and hollow drainage.

Subsurface drains are usually either pipes (AG pipe), mole drains, or a combination of both. Mole drains are subsoil drains installed in clay soils by drawing a mole plough through the soil and creating a channel. They will not work on soil with <35% clay or dispersive soil as they will collapse.

Raised beds are made with a bed-former that hills up beds about 2m wide. This raises the level of the planting surface and leaves open drains on either side of the beds to channel water out of the paddock.

Surface drains are usually considered frst where they can re-direct surface water that is exacerbating waterlogging. Common options include trench drains, spoon drains, and hump and hollow drainage.

Subsurface drains are usually either pipes (AG pipe), mole drains, or a combination of both. Mole drains are subsoil drains installed in clay soils by drawing a mole plough through the soil and creating a channel. They will not work on soil with <35% clay or dispersive soil as they will collapse.

Raised beds are made with a bed-former that hills up beds about 2m wide. This raises the level of the planting surface and leaves open drains on either side of the beds to channel water out of the paddock.

Land Drainage for Farming in Tasmania has more information on drainage options.

SECTION
4 I

Other soil constraints

Other soil constraints that affect productivity include:

  • Soil erosion – see more Module 3 and Module 6
  • Low fertility – see more Module 3
  • Biological constraints such as pathogens and pests. Seek agronomic advice for each specific challenge. Such constraints can impact long-term land values and therefore may be important to address quickly.

Resources and References

menuchevron-down