Table Of ContentsNext Page

Harvesting Native Grasses - Innovation and Development in the Central West

Andrew Briggs

Roadside Conservation Officer, The Rural City of Wangaratta, Victoria, 3677.
Former Native Grass Harvester, Native Grasses Inovation and Development Program, Department of Land and Water Conservation, Wellington, NSW.


In recent times there has been a surge of activity relating to the recognition and development of native grasses as valuable, sustainable entities to both production agriculture and habitat conservation/restoration. An important contributing factor in this recent activity has been research (both private and government) into the harvesting and sowing of native grass seed, and the development of a native grass seed industry.

One of the most significant contributors to the development of harvesting and sowing activity was the Natural Heritage Trust funded “Native Grasses Innovation and Development Project” (DLWC). A key focus of this program was to develop machinery that can efficiently and economically harvest and sow a wide range of perennial native grasses.

This project worked closely with the Barney’s Reef Landcare Group and Stipa and since the closure of the project this groups has continued to be very active in pursuing the continued development of this technology, and taking the next bold step: commercial and contract harvesting of native grass seed. This paper will provide a background of some of the most up to date native grass seed harvesting technology, and follow the progress of the continued pursuit of improvement.


If there is to be wide-scale acceptance of the use of native grasses in agricultural Australia, there are a number of critical obstacles, which must first be overcome. Of highest priority is the problem of seed supply. The harvesting of native grass seed has, until recently, been a slow and laborious task, usually resulting in small amounts of seed. Such efforts have been useful mostly on a small scale, and the cost of seed gathered in such ways restricting its use on a landscape scale.

Firstly, research has been hampered by the lack of readily available, good quality seed. There is therefore a relative lack of the type of solid information that is needed to persuade farmers to adapt to new species and technologies.

Secondly, re-establishment of native grasslands in a broad acre situation has not been generally successful, even where farmers are inclined to give it a go:

  • Use of seed with low levels of germination (often bought untested) has resulted in unsuccessful or poor establishment and has given the impression that it just can’t be done.
  • Lack of availability of the most appropriate species or provenance type has led to inferior or unsuitable species being sown with poor results.
  • The high cost of the seed that has been available has meant that it has been put out of the reach of the general landholder.

Thirdly, the same high cost, lack of reliable supply and most importantly, quality of seed, is a huge stumbling block for the fledgling native grass seed industry. It is also fair to say, however, that there is a huge interest in what native grasses have to offer in terms of sustainable grazing, habitat restoration or aesthetics. The established seed industry has effectively failed to take up the challenge and has been loath to invest significantly in order to overcome these problems.

Clearly, there was a need to develop equipment that could practically and economically harvest large quantities of seed. The Native Grasses Innovation and Development Program had as one of its aims therefore, to remove this important barrier to adoption.

This project resulted in substantial improvements to the accessibility of native grass harvesting machinery as well as increasing the experience and expertise necessary to harvest high quality seed regularly and reliably from natural stands. Two different machines were evaluated: the “Grasshopper” brush harvester, and the “Scorpion” vacuum harvester. It was felt that both these machines fulfilled several essential criteria:

  • Both machines were entry-level machines with low initial cost but significant flexibility and development potential should this be desired.
  • Both appeared to be easily transportable, sturdy enough to negotiate fairly rugged terrain, and running and maintenance cost was low.
  • Both machines were easy to operate, making them suitable for groups of people with a broad range of skill levels as may be experienced in community groups such as Landcare groups.

Parallel to the needs of technological advancement, was the need to locate remnant stands of native grasses of a size and purity suitable for harvesting. The development of best management practices for such areas, with the objective of sustainable yields of high quality seed was also crucial. These stands had to encompass all areas within the region, to ensure that provenance selection for re-establishment could be successfully carried out.

Since the closure of the Native Grasses Innovation and Development Project the Barney’s Reef Landcare Group, in co-operation with Rosevale Engineering, has continued the development of the Scorpion Harvester concept. As a result, a new harvester was borne, the “Rosevale Reaper”. This new machine has incorporated the basic design concepts of the Scorpion to result in a machine that has far greater versatility due to ease of transport, vastly improves operator comfort and safety and maintains an effective means of bulk seed collection. These developments have effectively made the original Scorpion Harvester obsolete, whilst the “Grasshopper” technology remains effectively unchanged.

It should be noted that the research and development detailed in this paper is occurring in the Macquarie and Lachlan Catchments of NSW (loosely termed the Central West). However, the species being targeted, seasons and other factors can be essentially overlayed into North Eastern Victoria, or indeed any other area in Australia, provided that some level of local knowledge in respect to seasonal influence on flowering times and growth habit is obtained.

Background Evaluation and Development

Prior to the latest design modifications the “Grasshopper” and the “Scorpion” harvesters were subjected to extensive design modifications. This is particularly evident in the “Scorpion” which in its present form bears only superficial resemblance to the original machine. As with most machinery, this process of development is on going, and with each season there are constant redesigns (and repairs!).

Where the scorpion and grasshopper harvesters were tested alongside each other both harvesters gathered comparable amounts of redgrass (Bothriochloa macra) (table 1) seed and this would probably hold true for the majority of species. Harvesting data of sufficient detail for the “Rosevale Reaper” was unfortunately unavailable at the time of writing, however an acceptable comparison can be made between this new harvester and the now obsolete Scorpion.

Harvesting efficiencies, ie. seed harvested compared with that available, range between 28% to 73% (Cole and Waters 1997) and seem comparable with those quoted elsewhere (Scholz 1995). However seed yields per hectare from unmanaged natural stands is relatively low ranging from 1.3 to 10kg/ha (Cole and Waters 1997). In practice, the relatively large areas offset this, and the low cost of production afforded by the largely opportunistic harvesting.

Table 1 Side by side peak harvesting performance of the “Scorpion MkI” and “Grasshopper” seed harvesters on redgrass (Bothriochloa macra) harvested in the Central West in 1997.


Average weight of 100 florets

Average no. of seeds in 100 florets

% unripe heads in harvested material

Total harvested material per plot (kg)











More detailed information on the development and practical usage of the Grasshopper, the Scorpion and the Rosevale Reaper will now be provided.

The “Grasshopper”

The Grasshopper harvester, a simple yet mainly conventional brush harvester, as described by Cole & Waters (1997) employs nylon brushes (similar in size to those of a road-sweeper) of varying coarseness, to literally “brush” the seed off the stems of the plant as they ripen. The detached seed is then carried into a hopper via the airflow generated by the brush. For a given species, finer or coarser brush types can be employed to maximise success. Additionally, the speed and height at which the brush operates can have enormous impact on the quantity and quality of the seed sample.

The “Grasshopper” weighs about 500-750 kg and attaches to a standard tow ball hitch on any 4WD vehicle and towed as one would tow a trailer. When harvesting, the machine is then swung out to the side of the vehicle, to facilitate harvesting seed that has not previously been passed over by the towing vehicle. This is particularly important when harvesting species of grass whose ripe seed require only minimal disturbance to be removed from the seed head, e.g. Danthonia spp.

For some other species, e.g. Chloris truncata, which have a low growth habit, it is not essential to tow the harvester to the side. Similarly, in rough terrain, or in the presence of many obstacles such as trees the harvester can be used in line with the vehicle.

The brush is belt driven, with power supplied by a 5.5 hp petrol motor. The machine is easily serviced, with a minimum of moving components, which are all easily accessed. This factor is very important if the harvester is to be used by individuals or a collective of landholders. Raising and lowering of the brush height is carried out either manually or via hydraulic rams which can be fitted optionally, and are operated remotely from within the cabin of the vehicle. Such an addition includes the ability to stop and start the brush automatically and makes the operation of the “Grasshoppers” far easier.

One of this harvester’s strengths is its versatility. It has been found to be effective on a wide range of species including those that have light fluffy seeds such as Danthonia spp., Dicanthium sericeum and Bothriochloa macra.

It also works quite well on species such as Paspalidium spp. and Chloris, where the fingers of the brush neatly strip the seed from the stem. It can also be used with reasonable success on kangaroo grass (Themeda australis) where the coarse brush is used to remove the whole floret from the stem.

At present the main limitation is with very low or lodged crops of Warrego grass (Paspalidium spp.) or with very tall stands of plains grass (Austrostipa aristiglumis) which can grow as tall as two metres.

The Grasshopper retails for around $6,500 - $8,500 depending on which version is selected and what optional extras are included.

Figure 1, the “Rosevale Reaper” at the First National Stipa Conference “Better Pastures, Naturally”, Mudgee, 2000.

The “Scorpion” and the “Rosevale Reaper”

This harvester was been developed by the DLWC in conjunction with the Barney’s Reef Landcare Group and Rosevale Engineering, and has since metamorphosed into the “Rosevale Reaper”. Whilst these two machines are no longer really comparable, in terms of the actual harvesting function, the system remains essentially the same. The harvesting head employs either brush or solid steel beaters of approximately 200mm diameter and a partial vacuum generated by a large air blower picks up the seed.

The main problems encountered with the original Scorpion included the reliability of the drive mechanism, operator comfort, lack of effective braking systems and difficulties with transportation over long distances.

Rather than attempt to rectify these problems on the existing machine, the self-propelled concept it was abandoned in favour of a demountable, utility mounted version (Figure 1). Thus, the “Rosevale Reaper” was born.

This development has had the added advantage of substantially reducing the size and cost of the machine (not including the purchase price of the utility) whilst increasing its overall reliability, operator comfort and safety.

The power source and hoppers of this harvester type mount onto the rear of a 4WD utility, while the demountable harvesting shroud attaches to the front of the vehicle via the bull bar. The beaters are driven with 12 Volt motors. As the vehicle travels forward through the grass stands the seed, or seed heads, are removed from the plant and transported via large flexi-hoses to the hoppers. A large fan at the rear of the vehicle provides the suction. Gauges measuring beater speed and fan speed make fine tuning the use of this harvester to the individual needs of the species being harvested possible. The seed bins are emptied straight into wool packs or similar bags (Figure 2).

The shroud at the front of the vehicle can be easily detached to allow safe transport from site to site. Raising and lowering of the shroud is achieved by either utilising a winch, or 12 Volt electric motors. The ability of the shroud to be lifted well clear of the ground, or alternatively lowered to just above ground level, means that it is able to harvest very tall species as well as very low growing species.

This feature makes this machine extremely versatile and in some circumstances it may produce cleaner seed samples than the “Grasshopper” due to the ability to lift the harvesting brush/beaters clear of the bulk of the “trash” along the ground.

Experience in the operation of the original “Scorpion”, despite the obvious shortfalls proved that the design concept of the harvesting head was sound, if in need of some refinement. The Rosevale Reaper had a fairly thorough working during the 2000/2001 summer, and for the first time harvested some stands on a contractual basis rather than for research or experimentation.

A number of species were targeted, including Microlaena stipoides, Themeda australis, Dicanthium sericeum, Bothriochloa macra and Danthonia spp. Harvesting rates were $95 per hour plus travel. One of the most successful efforts was the harvesting of over 200kg of Themeda australis florets from approximately 70 acres near Cessnock. The time taken to harvest this area was around 16 hours.

Reports from the owners and operators of the machine would seem to indicate that it was a highly successful testing of the machine, and that it has potentially established its place in the commercial native grass seed harvesting industry.

The Rosevale Reaper sells for around $17,000

Figure 2. Emptying the seed bins of the Rosevale Reaper.

Other Harvesting Issues:

A note on “Purity”

Purity, in reference to a sample of seed is usually defined according to a number of key points:

  • Inert matter contained within the seed sample
  • Presence of other, desirable seeds
  • Presence of weed seeds

Due to the traditional, monotypic nature of many commercial seeds, a common perception of purity, relating to a seed sample, is that there must be one and only one species of plant represented. This has been a perceived difficulty in marketing native grass seed; how does one get a “pure” sample of Microlaena, for example. In order to do this, one must be committed to removing not only large amounts of trash, but also large amounts of seed from other native species.

And yet, if one considers the nature of native grasslands, the extremely complex and diverse network of species, why would one seek to fragment this mixture. After all, it is in the diversity of a pasture community that lies strength; strength to resist and to recover quickly from upheavals such as drought or fire. The quality of a “pure” sample of native grass seed, therefore, might be better judged by the overall percentage of weed seed contained within it, not simply on the percentage of a dominant species.

Seed harvested from remnant grasslands are snapshots, taken from the environment to be re-established elsewhere. To attempt to establish only one particular species, when one might replace an entire community, is to travel down that path which was taken when we first began introducing pasture species. We would be weakening the very strengths that are integral in our native grasslands.

Location of Harvestable stands

The location of natural stands of grassland that are of a viable size for harvesting is of understandable importance. Within the central west catchment, these areas are difficult to find, however, depending on the species being targeted, are by no means rare. Extensive areas of such species as Red Grass (Bothriochloa macra) or Kangaroo Grass (Themeda australis) still exist in many areas.

The actual activity of physically scouting out areas to harvest can be very time consuming. With this in mind, it is beneficial to have two persons working together at harvest time; one seeking out areas to harvest, the other following with the harvester. Another reason for having two people on the ground is monitoring stands once they have been located; native grass seed can ripen and fall from the head extremely quickly, and without careful monitoring of both the plant’s development and climatic events harvesting opportunities can easily be lost.

What is of particular importance is to find harvestable areas of all species, across a wide range of their natural distribution. For example, large stands of Kangaroo Grass are common throughout the upper parts of the catchment, however are less common throughout the lower parts. This is of particular importance; when considering the re-establishment of harvested seed, one may then select seed that will potentially be best suited to the conditions of establishment.

When beginning to seek out these islands of grass, in a sea of altered and degraded landscapes, there are certain approaches that one can take which can be of great assistance. Firstly, areas where there has been minimal, or controlled disturbance have the potential to contain remnant stands. Such areas include cemeteries, Travelling Stock Reserves (TSR’s), country racecourses or even roadsides.

Of high importance, however, is the location of remnant stands on private land. In particular, land where there is an existing agricultural practice occurring, which contains remnant stands is of huge importance. In such areas, the presence of a grass remnant indicates, to a certain degree, that the prior management of the site has been in favour of the native grassland. From such discoveries, it may then be possible to develop or improve management strategies.

Locating remnant stands on private land is of enormous importance, not only do these areas often contain some of the largest and healthiest remnants, it is a vital tool in promoting the usage of native grasses. The acknowledgement and encouragement of those landholders who are responsibly and sustainably managing their grasslands achieves two main objectives. Firstly, it promotes the continued and enhanced usage of native grasslands, and secondly, it provides a valuable reward to the landholder who may for some time have been acting against the advice of his/her peers in respect to supposed “pasture improvement”.

Management of harvesting areas

The management of stands for harvesting is an area that is only really just beginning to be explored. Often, the quality of harvested seed has been compromised due to the presence of weed seeds, or large amounts of trash. On occasion, these items have been able to be removed as simply as sieving the sample to remove smaller, heavy seeds, e.g. Paterson’s Curse (Echium plantagineum) from Wallaby Grass Danthonia spp. At other times, however this is not so simple, for example the removal of Saffron Thistle (Carthamus lanatus) sepals from Redgrass (Bothriochloa spp.) seeds.

Strategies, which have been observed to have some benefit to preparing native grass stands, include slashing, burning, grazing and spraying, or combinations of these activities. The aim for such activities is to remove annual grasses and broadleaf weeds whilst still in a juvenile form. Such practices are usually employed in the spring or early summer, around one-two months in advance of harvest time. This is to ensure that there will not only be little or no weed seed contaminating the harvested seed, but also to reduce the amount of trash contained within the sample.

The ultimate in stand management is to create a carefully planned and adhered to regime, which includes an assortment of activities, applied at strategic times, over a long period. An example of a long-term practice, which has succeeded in some areas, is a “pulse grazing” system, which incorporates intensive, targeted grazing interspersed with long rest periods.

It must be noted that at virtually no two sites will the management regime required be exactly the same, and that incorrect management could have a serious and deleterious effect. Management needs to be tailored according to not only the species to be harvested, but also to whatever minor, yet important components that may exist.

Harvest preparations

Being prepared for harvesting is of the utmost importance if one is to successfully harvest quantities of seed. Such preparation moves beyond simply locating and/or managing stands.

Of equal importance to the overall success of harvesting are:

Seasonal effects. Every season is different, inducing vastly different responses from plant communities than what may have been previously observed. This means that not every species desired may be available to be harvested every year. One needs to be somewhat opportunistic; if an opportunity arises to harvest a species that may not be of high importance now, but may be useful later, it should be seized immediately.

Warehousing seed. When good seasons do occur, it is wise to harvest a certain amount of seed in excess of what the immediate requirements are (without being greedy). Stockpiling this seed will then provide a safety net if favourable conditions do not present themselves again next year, or for a number of years, for that particular species.

Storage facilities. Having appropriate storage facilities for the warehousing of seed is critical, or much hard work can be undone. Facilities need to be dry, vermin proof (or the seed stored in vermin proof containers), and of a relatively stable temperature, neither becoming too hot or too cold. Storage drums which have been found to be useful include large, 200-litre plastic pickle drums, or for smaller samples, 20 litre paint drums. Containers need to be clean and dry, although not necessarily sterilised.

Seed drying. The harvested seed needs to be dried immediately, particularly during the warmer months. Seed left unattended in woolpacks, or other bags, will quickly begin composting, thus effectively destroying most, if not all, of the seed. This factor needs to be carefully considered and arranged for prior to undertaking any harvesting activities. Ideal drying conditions will consist of a large area which is protected from weather, some airflow without being excessively draughty to prevent seed being scattered and vermin proof. Additionally, tarpaulins, or bin liners etc. must be purchased to spread the seed onto; this helps keep seed lots together and facilitates easy pick up of the seed once dry. Buildings such as shearing sheds are often ideal. The area requirement for drying seed must also be fairly large. For example, a woolpack around two thirds full of seed requires and area of no less than ten square metres and the seed must be spread no thicker than around fifteen centimetres. Once spread, the seed must then be turned daily to ensure even drying. Under ideal conditions, depending on the original moisture content of the seed, drying will take approximately one week.

Labelling. All seed must be labelled immediately upon collection. Labels must also accompany seed as it is being dried, and once again during storage. Accurate record keeping ensures that seed lines do not become mixed, and allows an assessment of such issues as establishment success relating to provenance selection, seasonal influence on seed quality and so on.

Equipment reliability. All equipment to be used must be checked to be free of faults before harvesting begins. For a harvester to break down, even for a couple of days at a crucial time could mean that an opportunity to harvest a particular species is lost. With this in mind, it is essential that the operators familiarise themselves with the piece of equipment to be utilised so that if minor breakdowns occur they can be fixed immediately. It is wise to carry spare items such as spare tyres and drive belts etc. whilst in the field. Also of great importance is vehicle maintenance, particularly in respect to air filters and radiators, which can quickly become clogged with seed and other matter. If this happens, particularly given the time of year, engines can quickly overheat and lead to long delays and costly repairs.

Timing of harvesting activities

Timing of harvesting activities is extremely important, and requires close monitoring of the stand to be harvested. While for many species ripe seeds can be found almost year round (depending on seasonal effects), there are definite time frames, which need to be matched to maximise the viability of the seed, being harvested.

Anecdotal evidence suggests that some species, for example Themeda australis, produce the bulk of florets during early summer and ripen around the last week in December to the first week in January. Later in the summer subsequent sets of florets may appear, however with a much reduced germination rate than the earlier seedings.

A table outlining approximate time frames for the harvesting of the main species available throughout the upper and mid Central West Catchment is provided below, however it may also be used as a general guide in other areas, including north Eastern Victoria. These times are only offered as a guide as seasonal variation may cause different or unusual behaviour.



Themeda australis


Chloris spp.


Bothriochloa macra


Dicanthium sericeum


Paspalidium spp.


Enteropogon acicularis


Danthonia spp.


Microlaena stipoides


Elymus spp.


Environmental Impact of harvesting

The environmental impact of harvesting activities is an important consideration. The two harvesters, which are employed to collect native grass seed, have been designed with “treading lightly” in mind, and only remove the ripest seed from the seed heads.

Unlike conventional harvesters, which physically remove a significant portion of the plant, the “grasshopper” or “Rosevale Reaper” harvesters simply knock the seeds out of the seed heads. This factor, combined with the differential way in which native grass seeds mature on any given plant, the actual amount which the harvester removes on any given harvest has a maximum success of around 60%, and is often significantly less.

Thus, at the time of harvest, the plant has already dropped a proportion of its seed, and the remainder is left to fall at a later stage. In addition, all species targeted for harvesting are perennial; therefor the survival of the individual is not necessarily dependent upon copious seed fall.

Even so, due to a desire to have as little impact as possible on this precious resource, harvesting sites are not usually targeted in consecutive years. This is to ensure that any depletion of the soil seed bank, and any effect of “trampling” from car tyres is minimised.


The success of native grass seed harvesting activities in recent years has advanced significantly. The community now has available to it machinery from this and other projects that have been developed and extensively field tested to a point where, season permitting, high quality seed can be harvested reliably and efficiently.

Furthermore, at least one collective of interested landholders, with assistance from the DLWC has successfully moved into the realms of commercial seed gathering. At this point in time there is also an accumulated depth of knowledge and experience relating to not only the actual activity of physically harvesting seed, but also on the many associated activities.

It is now, therefore, in the hands of the community, the seed industry and various other potential users of native grass seeds to continue to take up and make use of this technology, information and support base.

The ongoing challenge is to provide the seed products needed by land managers, at a price they can afford, to ensure their use on a landscape scale so that that our fragile ecosystems are returned to a more sustainable productive base by the use of our valuable native grasses.


Mr Ian Cole (DLWC, Centre for Natural Resources, Research Centre, Cowra, N.S.W.), for his input into the information contained within this paper relating to the “Grasshopper” harvester.


Cole, I.A. and Waters, C. (1997). Proceedings of the Twelfth Annual Conference Grasslands Society of New South Wales (Eds. A. Bowman and D. Michalk), pp. 95-103.

Scholz, G. (1995). A practical guide to rangeland revegetation in western NSW: using native grasses. Technical Report No. 33, NSW Department of Conservation and Land Management.

Previous PageTop Of PageNext Page