Overstory #125 - Direct Seeding
Sowing seeds directly onto prepared ground can be a cheap and effective method for establishing trees and shrubs over large areas. In fact, direct seeding can be up to 10 times cheaper than planting for broadscale revegetation.
The establishment of plants by this method is largely controlled by climatic conditions, soil type and weed competition. These factors have a significant bearing on soil moisture, which is vitally important to germination and early survival of seedlings in the field. Seedlings must be protected from grazing by vermin and livestock until they are beyond browsing height.
Other factors that play an important part in determining the success of the method include:
- correct choice of species, e.g. avoidance of frost-sensitive species on a site prone to severe frosts;
- ground preparation to provide a suitable seedbed;
- use of good quality, viable seed;
- correct seeding rates to ensure the required density of seedlings -- too little will lead to disappointment and too much will necessitate extra work for thinning;
- sowing when soil moisture is favourable for seedling germination and establishment; and
- using the correct sowing procedure so that seed is not buried too deeply.
A willingness to experiment is essential when direct seeding, as no one method will suit all situations.
The techniques described have been developed specifically for farms and include procedures to deal with competition from crop and pasture species. All aspects of the operation can be carried out using widely accepted farming practices and readily available farm machinery.
Influence of climate, site conditions and species
Like most farming, direct seeding is a game of chance. The chances of success are determined by the local climate, site conditions, selected species and available techniques.
The amount and pattern of rainfall are very important to the germination and establishment of seedlings in the field. Where rainfall is low and irregular it is possible to get false germination of seeds. This occurs when seeds begin to germinate after rain but are not able to survive due to a lack of follow-up rains and therefore inadequate soil moisture. Direct seeding has been very successful in the wetter winter-rainfall region of Australia (e.g. the Western District of Victoria) due to the relatively reliable rainfall patterns and high rainfall. However, in the arid zone and in the drier parts of the winter-rainfall zone (e.g. marginal farming areas in southern Australia) the method has had mixed success due to the less reliable rainfall patterns and low rainfall.
Where weeds are difficult to control it is best not to use direct seeding to establish trees. Poor results are invariably obtained with this method where there is dense weed competition. The introduced grasses and broad-leaved weeds compete aggressively for soil moisture, nutrients and light.
It is important when selecting species to consider not only the potential of the site and the suitability of the species for the site but also the suitability of the species for direct seeding. Some species (e.g. acacias) are well suited to direct seeding. They produce large quantities of seed that are easy to collect and they establish readily from seed sown on site. Species that cannot be propagated readily from seed will need to be established by planting seedlings that have been raised in a nursery.
Site preparation involves weed control, fencing, vermin control around the planting site and ground preparation to provide a suitable seedbed. It must be stressed that without due care with the initial step of site preparation it can be a waste of time and money to sow the seed.
Some weed control methods incorporate ground preparation. For example the fine tilth created by cultivation is ideal for sowing seeds provided the surface is firm and level. Where chemical means of weed control are used, ground preparation may be required to prepare a suitable seedbed just before sowing.
The time needed to achieve adequate weed control before seeding depends on the type of weeds present and the means of control used. Where persistent perennial weeds are present, chemical control may be necessary over two seasons before ground preparation and sowing can begin. The removal of weeds well before sowing conserves soil moisture.
However, on sites prone to erosion, complete removal of all grasses and other weeds should be avoided. On these sites weeds should be removed in strips or niches (spots) so that the areas cleared for sowing are relatively small. The weeds left adjacent to the cleared areas also provide some protection for the young seedlings.
The use of fertilisers and sowing of improved pastures have made direct seeding of native species more difficult on farms. The introduced pasture species are able to establish quickly on exposed, high-nutrient soils, suppressing or even preventing growth of the slower growing native species. Weed control has become essential in rural areas for successful direct seeding. The method employed will be influenced by site characteristics, such as soil type and weed spectrum, and the machinery available. A variety of approaches can be used. Mechanical methods include cultivation and grading (scalping); chemical methods use a knockdown herbicide, or a mixture of a knockdown and a residual herbicide. Mechanical and chemical methods may be combined, e.g. cultivation followed by the use of pre- or post-emergent herbicides.
It has been found that cultivation, with or without pre- and post-emergent herbicides, controls weeds for only a month or two before they begin to reoccupy the site. Grading away the uppermost topsoil substantially improves early seedling survival as it removes the weed-seed reserve and therefore effectively retards weed reinvasion. Removal of the topsoil does not appear to affect growth of the seedlings.
Opportunities for weed control after sowing are limited to those situations where herbicides can be applied selectively or where herbicides that kill the weeds but not the seedlings are available.
Although cultivation has generally given poor results, it has been relatively successful at one site where the soil has good moisture-retaining properties and weed competition is limited. However, where weed competition is a problem, methods that control weed reinvasion for at least a year after seeding must be used for direct seeding to be successful.
Sowing the seed
The seed may be treated before sowing to overcome dormancy, to hasten germination or to provide inoculation. Sometimes seed is pelleted for easier sowing or to protect the seed from insect predation or fungal infection. Pelleting is recommended only if exact prescriptions are known for the species being used because fungicides and other chemicals are poisonous to germinating seed, even at low concentrations.
Although pelleting has been found to improve seedling establishment in forests, there is little conclusive proof at present to support its use for a wide range of species on farms. It is regarded by some as an extra process complicating an otherwise simple operation.
When using only small-seeded species (e.g. eucalypts, bottlebrushes (Callistemon spp.)) rates vary from 0.1-- 1.0 kg/ha. However, when including large-seeded species with low viability (e.g. native pines (Callitris spp.)) a higher rate is required (e.g. 1.0-- 2.0+ kg/ha). High seeding rates add an unnecessary cost to the operation and can usually be avoided if the correct site preparation and seed treatment have been carried out.
Assessments of the quantity of seed required are generally based on the experience of the operator. However, it is possible to calculate the necessary amount of seed using the following equation:
# plants required/ha
kg of seed required = --------------------------------------
# viable seed per kg x % establishment
The viability of each seedlot in the seed mix should be determined. There are several techniques for testing seed viability. The simplest way is to take a sample of the seedlot and cut through each seed to determine the number of viable seed per unit weight. The contents of viable seed are usually white and fleshy.
Another method for determining seed viability is to germinate samples of seed at a temperature approximating field conditions at the expected time of sowing. The number of germinants is counted over a period of 3--6 weeks.
Conservative estimates of percentage establishment for small-seeded species would be 1% and for large-seeded species 5%. Underestimating the percentage will result in a greater density of trees or shrubs than is required. Percentage establishment figures for eucalypts are usually 1 - 5% though figures as high as 38% have been recorded. Similarly, figures for acacias are usually 5 - l5% but again much higher figures (up to 65%) are not uncommon. Results vary widely depending on the species and site conditions and are generally difficult to predict.
Time of sowing
The time of sowing is determined by rainfall patterns, likely occurrences of frost, optimum germination temperatures of the species used and weed problems at the site.
In Victoria and southern parts of New South Wales seed is sown in spring--early summer as soil temperatures are rising and weeds are dying. However, rainfall must be adequate to allow sufficient root development or the seedlings will not survive over the ensuing dry summer months.
Throughout most of the South Australian agricultural area spring rains are not reliable. Therefore seed must be sown as soon as possible after the break of season, generally in May--June, once weed control has been completed. Seedlings are then growing at the same time as a wide range of weeds, which can reduce seedling survival.
In the tropics, best results have been achieved when seed is sown at the beginning of the wet season, even though operations can be delayed at this time of the year by bad weather.
Optimum germination temperatures for species must be considered (Boland et al 1980). Species that favour low temperatures should be sown in winter and those that favour higher temperatures in spring-summer if rainfall patterns allow.
Where residual herbicides have been used for weed control, seed placement should aim to avoid contact with the chemical, particularly if it is likely to impede or prevent germination. Seed can be sown in riplines where the treated topsoil has been put to one side, or, if the site is stony and cannot be ripped, seed can be sown in niches. Niches are created by pulling away the topsoil so that a shallow saucer-shaped depression is formed.
Seed is placed in the base of the depression where the soil has been lightly disturbed. This technique can also be used to simulate grading on a small scale in rocky areas. Spacing between riplines or niches should be adjusted to suit the species sown.
Where non-residual herbicides (e.g. glyphosate) have been used, seed can be randomly spread over the surface. Although this method can be wasteful of seed, it does ensure an irregular placement of trees and shrubs and therefore a more natural appearance.
The equipment used to sow the seed should apply the right quantity of seed per unit area and distribute the seed correctly (e.g. not bury it too deeply). Seed can be sown using agricultural seeders (e.g. combine, drill, air seeder), by hand-operated spreaders (e.g. chest-mounted fertiliser spreader) or by hand.
In salt-affected areas seed is generally sown on raised furrows (mounds). Conversely, in sandy soils seed is sown in depressed furrows to reduce the effects of sand blasting.
Specialised machines have been developed to complete the different steps of direct seeding in one pass. These machines spray the ground, create a raised or depressed furrow along which a predetermined quantity of seed is sown at specified intervals with or without a mulch. The precision of the seed feeders ensures economical use of seed and better control of its placement, which is an advantage for shelterbelt and fodder-crop establishment.
Seed can be mixed with a medium to help spread it evenly. Sand and bran are commonly used. On new farmland in Western Australia and in mine rehabilitation work, fertiliser is often used as a medium. However, on sites where weeds are prevalent the use of fertiliser should be avoided. Fertiliser serves to enhance weed growth and any benefits to the sown species can be lost due to increased weed competition.
Moisture retention around the seed is improved and ant and/or insect predation can be reduced when some seed protection is provided. Optimum cover is generally considered to be 1--2 times the smallest diameter of the seed. A light soil covering can be achieved using peg harrows or dragging a chain over the site after sowing. Conventional tyne harrows can bury seeds too deeply and need to be carefully adjusted. Seed can also be compacted into the soil using a roller. Although results have been variable with compaction, it is generally better than no protection at all. Mulches can also be used to cover the seed lightly.
Insect damage and seed predation can be controlled by treating the sown area with insecticides but these chemicals should only be applied if there is no other alternative. Success has been achieved on farms throughout southern Australia without the use of insecticides, but it may be required where insects are a problem.
A site should be visited regularly after seeding to check the progress of seedlings. This allows any plant diseases or insect infestations to be treated promptly and may reveal reasons for failures.
Seedlings established by direct seeding should require minimal aftercare provided weed control before seeding was adequate to suppress weed reinvasion for the first year and the species have been sown at the correct seeding rates. By the end of their first year the seedlings should have a well-developed root system and further weed control should not be necessary.
If percentage establishment is higher than expected, thinning may be warranted. Unwanted seedlings can be killed by touching them with a ropewick impregnated with glyphosate.
It is important to keep a record of the materials and methods used and to monitor the site regularly during the first year. When results are poor it is easier to work out what went wrong if good records have been kept. There is no point in repeating the same mistakes the next year when simple corrective measures could substantially increase survival rates.
This article was adapted with the kind permission of the author and sponsoring organization (CSIRO) from:
Venning, J. 1990. 'Direct Seeding' In: Cremer, K.W. (Ed), Trees for Rural Australia, Inkata Press, Melbourne, Australia.
Boland, D.J., Brooker, M.I.H. and Turnbull, J.W. (1980). Eucalyptus Seed. CSIRO, Canberra.
Clemens, J. (1980). 'Direct seeding of native woody plants.' Landscape Australia, 4/80, 279--284.
Venning, J. (1988). Growing Trees for Farms, Parks and Roadsides. Lothian Books, Port Melbourne.
About the author
Julianne (Jackie) Venning specialises in environmental and natural resource management. She holds a Ph.D. in biological science and an MBA. Her earlier work involved the conservation and management of rural landscapes, particularly the retention, rehabilitation and re-establishment of remnant native vegetation in southern Australia. More recently she has been involved in the development and application of environmental auditing tools for monitoring and reporting on sustainable development. Recent publications include Towards Sustainability: Emerging Systems for Informing Sustainable Development (Venning, J. and Higgins, J., Eds, 2001, UNSW Press). Her contact details are: Dr Jackie Venning, Office of Sustainability, GPO Box 1047, Adelaide, South Australia. 5001 Australia. Ph: +61 8 8204 927, Fax: +61 8 8204 9144, Email: email@example.com .
Related editions to The Overstory
- The Overstory #91--Fodder Tree Establishment
- The Overstory #88--Revegetation Planning
- The Overstory #72--Microenvironments
- The Overstory #67--Commercial Timber Potential for Farm Forestry
- The Overstory #48--Farm Forestry agroforester.com/overstory/overstory48.html
- The Overstory #44--Integrated Systems Approach
- The Overstory #18--Designing Resource Systems