Gamba grass (Andropogon gayanus)

Assessment ID: 1066
Jurisdiction: North Territory
Methodology: Northern Territory Weed Risk Management System - terrestrial
Management area: Darwin Region

Notes: 

Status: PUBLISHED
Last updated: 27/3/2026 6:02:32 pm AEST

Questions answered: 39 / 47

Weed Risk: 485.37 (Very high)
   Invasiveness (raw/adjusted): 9.00 / 7.497
   Impacts (raw/adjusted): 18.00 / 9.468
   Potential distribution (raw/adjusted): 13.00 / 6.838

Feasibility of Control: 31.46 (Low)
   Control costs (raw/adjusted): 5.00 / 4.545
   Current distribution (raw/adjusted): 1.00 / 1.000
   Persistence (raw/adjusted): 9.00 / 6.921

Result: Targeted control (including biocontrol), protect priority sites

Invasiveness

A1. What is the ability of the plant to establish amongst intact native environments?
Answer: Very High

Comments: 

This mapping exercise clearly showed that roadsides and other disturbance corridors have been the major conduits for the spread of gamba grass in the NT. Originally planted for pasture on cattle properties, Gamba is currently spreading from these properties into surrounding areas, replacing native grasses (Kean, L. and Price, O. 2003a).

This summary paper emphasised that gamba is not restricted to these highly disturbed areas and it is capable of invading long distances (>500 m) from roadsides and pastoral properties into relatively undisturbed areas (Whitehead, P. J. and Wilson, C. 2000).

The results of this study showed that soil or canopy disturbance are not essential for the establishment of gamba grass, but both these factors will promote establishment and soil disturbance will increase survival in the first year (Setterfield, S. A., Douglas, M. M. and Hutley, L. B. 2005).

Observational data from this study showed that gamba grass can establish across a range of habitats, from wetland margins through to upland savanna. It is particularly common around riparian corridors (Flores et al. 2005).

The predicted potential distribution for gamba grass includes all savanna areas, wetland and rainforest margins north of Daly Waters (15°S); the represents an area of approximately 380 000km2 (Steve Wingrave, NT Weeds Branch, Pers Comm. 2005).
 
Gamba grass was reported in this study to have been observed to be colonising non-pastoral land in recent years (Barrow, P. 1995).

This study recorded a maximum spread of 7km at Wildman Reserve, resulting in the minimum estimate of the mean rate of spread in this area being 7km/21yr, or 333m yr -1. These measurements were obtained consulting historical records to ascertain dates of planting of gamba at specific sites and inspecting these sites (Barrow, P. 1995).

The rate of spread through native vegetation can be up to 110 m per year (Barrow, P. 1995).

A2. What is the reproductive ability of the plant?

(a) Time to seeding
Answer: 1 year or less

Comments: 

Gamba grass will set seed in the first year when fertilised but in an unfertilised situation, it will generally no flower in the first year because the plants are too small (Arthur Cameron, NT Pastures Branch, Pers Comm. 2005).

However, gamba plants have been observed to flower in the first year in undisturbed, unfertilised situations in the NT. Fertility of the seed in the first year is unknown but flowering and viable seed production will occur within two years (Setterfield & Douglas unpublished data).

(b) Annual production of viable seed /m2 or /plant
Answer: High

Comments: 

This study quantified annual seed production from mature gamba plants in one season in tropical environments in the Top end. Seed production varied depending on the number of inflorescences produced per plant which was an indication of the age of the plant. In the Top End, a mature plant has the capacity to produce up to 250,000 seeds in one season, with a potential viability of 46-65%. Statistics for seed production of Gamba grass vary depending on author and sampling methods. Estimates range from between 600 – 3,500 seed/m2 (Flores, T. 1999). 

Results from this study recorded average seed production of Gamba grass at 70 000 seeds/m2 (Flores, et al. 2005, Flores, T., Setterfield, S. A. and Douglas, M. M. 2005).

This management plan for gamba grass summarises existing literature and states that gamba grass can produce between 2 000 – 77 000 seeds m2. Production of seed is affected by climatic variables, soil fertility and plant density (Clifton, P. 2005a).

This study measured the seed rain of gamba grass and was found to be approximately 103 seeds m-2 yr-1. Unpublished data by the author found a mean of 324.4 (± SD 6.66) seeds g-1. These figures give an annual estimated seed production ranging between 15 – 130 kg ha -1 (derived from three separate sites). These results show that wild gamba grass (i.e. not sown) is capable of producing viable seed in quantities comparable to those produced under managed conditions (Barrow, P. 1995).

(c) Vegetative reproduction
Answer: None

Comments: 

Vegetative reproduction by tussocks has not been observed (Bowden, B. N. 1964).

A3. Do propagules of the plant have properties that allow them to be dispersed long-distance by natural means?

(a) Flying animals (birds, bats)
Answer: No

Comments: 

Yes. Gamba grass seeds are light and can become attached to feathers (Clifton, P. 2005a).

(b) Other wild animals
Answer: Yes

Comments: 

Yes. Gamba grass seeds are light, with hairs which enable them to become attached to animals. This is particularly the case for buffalo, cattle and pigs in the NT (Natalie Rossiter-Rachor, Charles Darwin University, pers. comm, 2005).

(c) Water
Answer: No

Comments: 

This management plan suggests that seeds may be transported by water but there is no scientific evidence to confirm this. “It is not known how far this fallen seed is transported by wind or water along the ground, although the movement of seed by water may be significant on slopes” (Clifton, P. 2005a).

In this thesis rain and wind are identified as primary vectors of seed although they are only responsible for transportation over short distances (Flores, T. 1999).

(d) Wind
Answer: Yes

Comments: 

Although gamba grass produces light and fluffy seed which can be spread by the prevailing winds during the main seeding time (May/June), and later in the dry season, as plants will flower if soil moisture is available, seed is generally dispersed only short distances by wind (Clifton, P. 2005a).

This study in the NT showed that in five months of sampling, very few seeds (<1%) fell more than 10m from the parent plant. The great majority (95.6%) fell within 5m of the plant. Although the seeds are adapted to a certain extent to wind dispersal, this is clearly not the major means of long range dispersal and cannot account for the rates or distances of spread previously observed in this study (Barrow, P. 1995).

In this thesis rain and wind are identified as primary vectors of seed although they are only responsible for transportation over short distances (Flores, T. 1999).

A4. How likely is long-distance dispersal by human-assisted means?

(a) Deliberate spread by people
Answer: Occasional

Comments: 

The primary mode of long-distance dispersal of gamba grass is through planting as a pasture plant. It has been widely planted since the 1980’s in the NT (Arthur Cameron, NT Pastures Branch, Pers Comm. 2005).

(b) Accidentally by people and vehicles
Answer: Common

Comments: 

This management plan for Gamba suggests that dispersal can occur along transport corridors. Seeds may be caught in passing vehicles, picked up and deposited during road making or corridor maintenance by machinery such as graders and water trucks; dispersed during corridor grass slashing or mowing; moved in soil (Clifton, P. 2005a).

This study found that 52% of cars entering Kakadu National Park carried grass seeds (Lonsdale, W. M. and Lane, A. M. 1990).

(c) Contaminated produce
Answer: Occasional

Comments: 

Seed may be transported in hay bales and also moved in soil being used for fill etc. (No reference).

(d) Domestic/farm animals
Answer: Occasional

Comments: 

Cattle may spread gamba seed, the seed attaching to the animals coat (Natalie Rossiter-Rachor, Charles Darwin University, pers. comm, 2005).

Impacts

B1. What is the plant's competitive potential?
Answer: High

Comments: 

At some locations, gamba grass has become a dominant grass species, especially ungrazed areas. While it is most abundant within cleared grazing paddocks, where it was originally planted, as well as nearby non-grazed, disturbed sites such as roadsides, there is little doubt that it is spreading into relatively natural (non-grazed and uncleared) woodlands. Visual observations by the author suggest that gamba grass is starting to invade Litchfield National Park, one of the Northern Territory’s major tourist attractions (Kean, L. and Price, O. 2003a).

The results of this study showed that Gamba grass invasion reduces tree recruitment with seedling density 75% lower in invaded sites compared to native grass sites (Clifton, P. J., Hutley, L. B., Setterfield, S. A. and Douglas, M. M. In prep.).

Changes in fire regime caused by gamba grass could also threaten the survival of rare plants such as cycads, plants that feature heavily in tourism promotions for the NT (Liddle, D. 2004).

Flame heights of gamba grass fires are much higher than native grass fires and are typically higher than the tree canopy. These fires could dramatically alter the structure of the native vegetation (Rossiter, N. A., Setterfield, S. A., Douglas, M. M. and Hutley, L. B. 2003).

This prediction is supported by a flame simulation model which suggests that gamba grass fires could kill most of the over-storey tree layer within a century (Cook, G. and Liedloff, A. Unpublished).

However, tree cover can be halved within 5 years where gamba grass invasion is coupled with poor fire management (i.e. late dry-season fires) (Setterfield, S. A. and Douglas, M. M. unpublished data).

More references available on request from weedinfo@nt.gov.au.

B2. What is the plant’s potential to modify the existing fire behaviour and alter the fire regime?
Answer: Significant potential

Comments: 

There is strong evidence that gamba grass is having an insidious impact on native vegetation by increasing fuel loads. Since grass grows to 4 m tall and produces up to 10 times as much biomass as native grasses (gamba grass typically produces a biomass of 11 – 15 t/ha, but up to 30 t/ha, compared to around 3 – 5 t/ha for native grasses), it produces a large amount of fuel for dry-season fires (Howard 2002, Rossiter, et al. 2003, Barrow, P. 1995).

Compared with sites dominated by native grasses, sites dominated by gamba grass have increased fire intensities by up to 8 times. Even early in the dry season, when gamba grass has only just started to cure, fire intensity is more than 3 times as high as that recorded in adjacent native grass savannas (Rossiter, et al. 2004, Rossiter, N. A., Setterfield, S. A., Douglas, M. M. and Hutley, L. B. 2003).

In the absence of gamba grass, early dry-season (May-June) savanna fires in the NT tend to be very low in intensity (c. 2000 kWm-1), patchy and limited in extent. Late dry-season fires are more intense (c. 8000 kWm-1) and extensive than early dry-season fires (Williams, R. J., Gill, A. M. and Moore, P. H. R. 1998).

This study measured fire intensities and concluded that fire intensities in savanna invaded by gamba grass can reach 24 000 kWm-1 measured in the early dry-season (Rossiter, N. A., Setterfield, S. A., Douglas, M. M. and Hutley, L. B. 2003).

More references available on request from weedinfo@nt.gov.au.

B3. What is the plant’s potential to restrict the physical movement of people, animals, vehicles, machinery and/or water?
Answer: High

Comments: 

When uncontrolled gamba often forms dense stands which restrict the movement of cattle (cows and buffalo), people and vehicles (Natalie Rossiter-Rachor, Charles Darwin University, pers. comm, 2005).

B4. What is the plant’s potential to negatively affect the health of animals and/or people?
Answer: High

Comments: 

There is evidence that bushfire smoke is detrimental to human health and can increase the incidence of asthma. Increased fuel loads in gamba grass areas will increase the amount of smoke in the atmosphere. Further research into the effects of bushfire smoke on human health is currently being undertaken (Johnston, F. H., Kavanagh, A. M., Bowman, D. M. J. S. and Scott, R. K. 2002).

B5. Does the plant potentially have negative effects on natural and cultural values?

(a) Reducing habitat quality for native animals?
Answer: High

Comments: 

With many Top End fauna species already in decline, widespread invasion by gamba grass and consequent altered and increased frequency and intensity of wild fire is likely to push many indigenous species further along the continuum towards threatened status or extinction. For example hollow dependent bird and mammal species and granivorous birds are likely to undergo further major population declines as gamba invades their habitat and removes their breeding and feeding resources (Woinarski, unpublished data).

Results from this study found a substantial negative effect of gamba grass on reptiles. Species richness of reptiles was significantly lower (on average 40% less) in savanna woodland areas of high Gamba cover (average 4.9 species) cf. zero cover (8.2 species) (Beggs, K. Unpublished).

(b) Threatened species or communities
Answer: More than one

Comments: 

Yes. The following species are listed as threatened under the EPBC Act: 

 -  Darwin palm (Ptychosperma bleeseri (macarthurii) endangered; The Darwin palm is only found in eight small rainforest patches east of Darwin, occurring in an area 30km long by 20km wide. Frequent fires prevent seedlings and young palms from growing while also allowing weeds to become established (Anon. 2003, Woinarski, J. 2003).

 -  Eastern partridge pigeon (Geophaps smithii smithii) vulnerable. Inter related changes in grass composition and fire regimes across much of the Top End threaten this species by reducing diversity of native grasses, which in turn changes the diversity, timing and abundance of seed available as food to the partridge pigeon. Increased fuel loads and the change in fire regime from small, patchy burns (Traditional Aboriginal) to the current regime of more extensive, hotter fires (Anon. 2003b, Kerrigan, et al. 2002a, Holmes, J., Bisa, D., Hill, A. and Crase, B. 2005).
 
 -  Gouldian finch (Erythrura gouldiae) endangered (No reference).
 
 -  Northern quoll (Dasyurus hallucatus) endangered (No reference).
 
 -  Yellow-snouted gecko (Diplodactylus occultus) EPBC case under consideration and listed a vulnerable under NT legislation. This species is endemic to the NT and is known from only a few locations. Hotter more frequent fires arising from gamba grass invasion are likely to increase the mortality of terrestrial geckoes and their eggs, to reduce the leaf litter in which they shelter and to increase predation risks. Further more almost all the records for this species are from Wildman Reserve in sites with moderate, low or no Gamba grass cover. If Gamba spreads further this species will suffer enormously (Beggs, K. (Unpublished)).
 
More references available on request from weedinfo@nt.gov.au.

(c) Sites of natural or cultural significance
Answer: More than one

Comments: 

B6. Is the plant presumed to have negative effects on environmental health?

(a) Soil chemistry/stability
Answer: Yes

Comments: 

Gamba grass affects a range of soil nitrogen pools and fluxes. For example it changes mineralisation rates, soil total N and the amount of N lost due to fire (Natalie Rossiter-Rachor, Charles Darwin University, pers. comm, 2005).
 
Could possibly affect soil stability. Field observations suggest that in areas where dense gamba occurs, patches of bare ground occur between gamba plants, this could result in increased run off and erosion etc. A similar phenomenon occurs in Buffel grass (Natalie Rossiter-Rachor, Charles Darwin University, pers. comm, 2005).

(b) Water quality
Answer: No

Comments: 

Unlikely. No evidence found to suggest that gamba grass affects water quality (No reference).

(c) Hydrology
Answer: Yes

Comments: 

Daily water use (in the wet season) is three times higher for gamba grass than native grasses. Preliminary data analysis shows that gamba grass invasion reduces the drainage of water through the soil profile and deep drainage was 20 to 50% lower in gamba invaded sites than in native grass sites. This will result in increased competition for soil moisture with woody species which will be particularly critical during the dry season when deep drainage provides a critical reserve of moisture sustaining evergreen tree canopies (Rossiter, N. A., Setterfield, S. A., Douglas, M. M., Hutley, L. B. and Cook, G. 2004).

Initial results from a study in progress suggest that gamba grass can reduce the amount of rainfall draining to depths of one metre or greater. Moisture at this depth is thought to be a vital source of water for evergreen trees to maintain transpiration throughout the long dry season. Therefore, reductions in deep-water drainage are likely to increase water stress and result in reduced health and reproduction of evergreen trees. Catchment scale infestations could also affect stream flow and potentially impact on wetland areas (Hutley, et al. Unpublished, Rossiter, N. A., Setterfield, S. A., Douglas, M. M., Hutley, L. B. and Cook, G. 2004).

Potential distribution

C1b. What is the Climatch suitability score (which indicates the proportion of the NT environment that is suitable for the plant)?
Answer: 7

Comments: 

Gamba grass is able to tolerate drought, fire and low nutrient soils, making it well suited to conditions in north Australian savannas. It requires an annual rainfall of at least 600 mm per year, although the current distribution is mostly north of Katherine (950 mm yr-1). It has been able to establish in Eucalyptus dominated open forest/woodland savannas; along creek-lines; floodplain fringes; disturbed vine thickets; and patches of Melaleuca open forest that aren’t seasonally flooded. It is adapted to most soil types except for heavy clays where water logging reduces survival. It is estimated that gamba grass covers an area of 10 000-15 000 km2, and is prominent in the Darwin, Palmerston and Litchfield Shires, the Coomalie Community Government, Adelaide River, Mary River, Douglas River and Lower Daly River regions (Figure 2.). Its potential range includes all savanna areas, wetland margins and rainforest margins north of Daly Waters. This represents an area of approximately 380 000 km2 (Clifton, P. 2005a).

This report states that although gamba grass is native to Africa, it now has a pantropical distribution. In Africa it is renowned for its drought resistance (Clifton, P. 2005a).

Gamba grass was planted experimentally at six stations in the NT located in the 600-1500mm rainfall zone and the authors found that gamba grass showed “excellent growth” at all sites sown. Soil types included red earths, glazed podsolic, with a high water table and some flooding, lithosol, sandy red earth clay loam and Lateritic podsolic. Therefore, gamba grass tolerates a very wide range of soil types (Cameron, A. G., Miller, I. L., Harrison, P. G. and Fritz, R. J. 1984).

C2. How many broad habitat types in the NT will the plant potentially naturalise in (up to 5)?
Answer: Two

Comments: 

C3. What is the potential of the plant to occur throughout its favoured habitat in the NT (as identified in question 2)?
Answer: Most

Comments: 

Control costs

A1. How detectable is the weed?

(a) Distinguishing features
Answer: Sometimes distinct

Comments: 

An erect, tussock-forming perennial grass to 4 m tall. Stems robust, often hairy. Leaf blades softly hairy with a prominent white midrib. Flower heads on thick stems held well above the leaves, consisting of loosely branched, hairy spikelets, giving a fluffy appearance. Flowers/fruits mainly Apr-Aug (Smith, N. M. 2002).

Leaf blades 30-60 cm long and up to 3 cm wide with a distinctive white midrib and covered with soft hair. These hairs hold droplets of water on the leaves after dew or rainfall, a trait that can help to identify the species (Clifton, P. 2005b).

Gamba grass plants are easily identified. The growth habit distinguishes gamba grass from other improved and native pastures (Lemcke, B. and Cameron, A. G. 1996).

(b) Active growth period
Answer: 4-8 months

Comments: 

Perennial grass (Csurhes, S. 2005).

It has the ability to remain green well into the dry season and provide a significant flush of early season growth at the beginning of the rainy season (Oram, R. N. 1990).

One of the major attributes of A. gayanus is its ability not merely to survive a long dry season of several months duration but also to remain green for much of it and to begin renewed growth very early in the following rains (Bowden, B. N. 1964).

Compared with native grasses, gamba grass..cures later in the dry season (Douglas, M. M. and Setterfield, S. A. 2005).

There is still, generally, live leaf material on gamba plants until the end of May (N Rossiter-Rachor, Charles Darwin University, pers comm. 2007).

(c) Height at maturity
Answer: >2 m

Comments: 

Reaches 1-4 m in height at maturity (Oram 1990).

A2. What is the general accessibility of infestations at the optimum treatment time?
Answer: Medium

Comments: 

If stands are large or plants are among trees, access is prevented and it may be difficult to eradicate the stand in the short term (Cameron, A. G. 2000).

Within its core area of infestation in the NT (Darwin, Palmerston and the Litchfield Shire), Gamba grass is most common on freehold and crown lease land and vacant crown land. It is significantly less common on conservation reserves and pastoral leases. It is most common in alluvial areas and drainage lines and around swamps and billabongs. Where it occurs along the road corridor it almost always extends beyond the corridor into existing vegetation (Kean, L. and Price, O. 2003a).

I suggest that gamba grass has a medium accessibility. Some areas of Gamba grass will be impossible to access, through most should be available at sometime before seeding (P. Jefferies, M. Fuller & G. McSkimming, pers comm. November 2006).

I would suggest that gamba grass has medium general accessibility, as you often need equipment such as quads to reach the grass in woodland areas (NT Weed Control Consultants, pers comm. November 2006, P. Jefferies, M. Fuller & G. McSkimming, pers comm. November 2006).

A3. How expensive is control of the weed in the first year of targeted control, for an infestation that has reached maximum weed density?

(a) Chemical costs
Answer: Medium

Comments: 

Total: $252 per hectare 
[Note: This would place Gamba grass control in the Medium ($100- 250/ha) chemical cost category].
 
Two primary treatments, each at a cost of: 
8.4kg Glyphosate Bi Dry + 2.4L Uptake and 1200L water per ha = ~ $84/ha 

One follow up treatment at a cost of: 
4.2kg Glyphosate Bi Dry + 1.2L Uptake and 600L water per ha = ~ $42/ha 

Total: $252 per hectare 
(NT Weed Control Consultants, pers. comm. November 2006)
 
$420 per hectare for tall dense infestation (total cost for operation and labour) (Roni Opden, NT Weeds Branch, pers. comm. 2006.).

For mission grass: 20 L of glyphosate is generally sufficient to treat 3- 10 hectares [i.e. between 2 - 6.6 per hectare]. This same chemical and ratio is also used to kill Gamba grass” [A 20 L drum of glyphosate at Landmark is between $119 and $130 (depending on the strength). This means that it would cost somewhere between $12 and $36 to treat a hectare of gamba grass] (Clifton, P. 2005a).

(b) Labour cost
Answer: High

Comments: 

Total: $525 per hectare 
[Note: This would place Gamba grass control in the Very High (>$500/ha) labour cost category]

Two primary treatments, each at a cost of: 
2 hr x 2 operators per Ha =~ $210 

One follow up treatment at a cost of: 
1hrs x 2 operators per Ha =~ $105 

Total: $525 per hectare 
(NT Weed Control Consultants, pers comm. November 2006) 

$420 per hectare for tall dense infestation (total cost for operation and labour) (Roni Opden, NT Weeds Branch, pers comm. 2006).

(c) Equipment cost
Answer: Medium

Comments: 

Equipment used: 
Quickspray 
[Note: This equipment would place Gamba grass control in the Medium equipment cost category]

Two primary treatments, each at a cost of: 
 1 x 4WD + QuikSpray @ 2 hrs = ~ $75 

One follow up treatment at a cost of: 
1 x 4WD + QuikSpray @ 1 hrs = ~ $37.50 

Total: $187 per hectare 
(NT Weed Control Consultants, pers comm., November 2006).

A4. What is the general community perception of this weed within the region?
Answer: Medium

Comments: 

Gamba grass is considered economically beneficial by some pastoralists in the NT. Young growth provides a palatable feed for cattle and when well managed can carry 40 times more cattle than native pastures. However it must be carefully managed and continually grazed or it will become unpalatable. There is mixed support for Gamba grass among pastoralists. Some recognize its risk as a weed and a fire hazard and its limited use in grazing. It is no longer recommended by the NT Gov. Gamba seed production is not a major industry (Csurhes, S. 2005).
 
Given that gamba grass is wanted by some land users but not wanted by others, effective and well coordinated management of the species is required to limit conflict between land user groups (Clifton, P. 2005b).

Pastoralists in the top end plant gamba grass on upland areas as trial results and industry experience has shown that it will feed and fatten livestock at high stocking rates during the wet season and carry up to 40 times more cattle than native pastures. It is a grass that requires a high level of management for it to remain productive and not to become a weed in other areas. It is not recommended for use on smaller blocks as it is very hard to manage. DBIRD is not recommending new sowings of gamba grass (Cameron, A. G. 2000).

Gamba grass is most common on freehold land and is often associated with drainage lines and wetter areas. The high prevalence on freehold land may reflect the dense road network and vehicle movement aiding dispersal. Freehold land tends to be under residential, industrial or horticultural use (Kean, L. and Price, O. 2003a).

Current distribution

B1. What is the current pattern of the species’ distribution across the weed management region?
Answer: Widespread

Comments: 

It is estimated that gamba grass covers an area of 10 000 – 15 000 km2. Its potential range includes all savannah areas, wetland margins and rainforest margins north of Daly Waters. This represents an area of approximately 380 000 km2 (Approximately 4%) (Clifton, P. 2005b).

It has been able to establish in open forest, woodland savannas, along creek lines, floodplain fringes, disturbed vine thickets and Melaleuca forests that aren’t seasonally flooded. It is estimated that gamba grass covers an area of 10 000 – 15 000 km2 and is prominent in the Darwin, Palmerston and Litchfield Shires, the Coomalie Community, Adelaide River, Mary River, Douglas River and Lower Daly River regions (Clifton, P. 2005b).

Gamba grass is mostly found on pastoral properties in an area bounded by the NT coast in the north and west, the boundary of Kakadu in the east and pine creek in the south, with small stands outside these boundaries (Cameron, A. G. 2000).

Most infestations occur in the pastoral region between Darwin and Katherine. The densest and most extensive infestations on non-pastoral land occur in the upper Adelaide River, Humpty Doo, Berrimah, Wildman Reserve and Mudginberri areas. These infestations consist of dense to scattered A. gayanus plants covering areas ranging in size from hectares to square kilometres (Barrow, P. 1995).

Persistence

C1. How long will it take to reach the maintenance period?
Answer: High

Comments: 

Resistant to fire, slashing and grazing. It has short rhizomes which are an adaptation that helps the plant produce new growth after the above-ground material has been destroyed. Most effective control is foliar application of glyphosate during the wet season (Csurhes, S. 2005).

Gamba can be controlled by spraying in the early wet season. Plants should be sprayed before may-June to prevent seeding. Early spraying makes it easier for herbicide to penetrate to the centre of the crowns of established tussocks. If plants are tall and dry mature material needs to be removed by slashing or burning to encourage regrowth which can then be sprayed. Follow up checking for 2 or 3 wet seasons is necessary (Lemcke, B. and Cameron, A. G. 1996).

Spraying plants while young reduces herbicide use but plants must be actively growing with leaves at least 40cm long to enable sufficient uptake by the plant. Plants sprayed at 40 cm can usually be effectively treated with a single pass with minimal off target damage. Larger plants need to be sprayed from all sides, are difficult to wet sufficiently and leads to off-target damage (NT Environment Centre. 2005).

Hand pulling and grading can be effective. Slashing and fire will remove growth but will not kill tussocks. Spraying with glyphosate herbicide in the wet season is effective if dry mature growth is removed by grazing, mowing or burning to encourage fresh regrowth which can be sprayed (Cameron, A. G. 2000).

This herbicide trial found that targeted control at the end of the wet season with optimal rates of Glyphosate achieved over 80% kill rate. This paper also cites unpublished results, also from P Barrow, where 1% Glyphosate 360 application using high volume spray equipment in the active growth period in the mid wet season resulted in a 100% kill rate (Barrow, P. 1995).

C2. What is the minimum time period for reproduction of sexual or vegetative propagules?
Answer: <1 years

Comments: 

Gamba grass is capable of producing seed in its first year (Clifton, P. 2005b).

Gamba will set seed in the first year when fertilised but in an unfertilised situation it generally will not flower the first year because plants are too small (Arthur Cameron, NT Pastures Branch, pers. com. 2005).
 
At a research study site (in undisturbed bushland) near Palmerston, gamba grass plants flowered when they were less than 1 year old. The plants flowered in late November. However, it is unclear if the seeds were viable (N. Rossiter-Rachor, Charles Darwin University, pers comm. 2006).

C3. What is the maximum longevity of sexual or vegetative propagules?
Answer: <2 years

Comments: 

Seed surviving in the seed bank after 12 months is negligible (<0.1%) (Flores, T. 1999).

Seed longevity of gamba grass was short, with rapid declines in seed viability in the dry season months. Seed viability for gamba grass declined from approximately 89% in June to less than 5% in December. The survival of seeds over the wet season is also extremely low, but some seeds did survive into the following dry season (Setterfield, S., Bellairs, S., Douglas, M. and Calnan, T. 2004).

C4. What is the threat of reinfestation from outside the management region?

(a) Long-distance (>100m) dispersal by natural means
Answer: Occasional

Comments: 

Gamba grass is dispersed by a range of vectors over a range of distances. Wind spreads seed short distances, water may spread seed significant distances on slopes, the movement of animals may also spread seed. Thermal air movements from fires area also thought to disperse seed off plants. It has been suggested that the rate of spread of infestations may be as high as 110 m per year (Clifton, P. 2005b).

(b) Long-distance (>100m) dispersal by human means
Answer: Occasional

Comments: 

Transport corridors act as major corridors of dispersal and can result in new outbreaks establishing many kilometres from known infestations. Seed may be caught in passing vehicles and deposited, especially by machinery such as graders and slashers, into disturbed areas that are highly susceptible to invasion. May also be transported in hay bales (Clifton, P. 2005b).

Seed can be spread in mulching hay that has been cut late, by vehicles or machinery, especially graders, mowers and slashers, or by animals. Most spread along roadsides is by maintenance machinery. Disturbance associated with earthmoving equipment increases establishment. In some instances gamba grass has established a long way from known stands (Cameron, A. G. 2000).

The primary mode of long-distance dispersal is through planting as a pasture plant (Csurhes, S. 2005).

Gamba is a highly productive and palatable fodder. Despite these benefits it is no longer recommended for new sowings by DBIRD. It is a difficult grass to manage and is avoided by stock if allowed to grow taller than 90 cm, thus requiring either de-stocking, slashing or burning. Poorly managed pastures can become a fire hazard. Gamba grass is not highly regarded as a hay species and has not been widely sown for this purpose. Some properties where gamba grass is not managed by cattle have baled it for hay (Clifton, P. 2005b).

Pastoralists in the top end plant gamba grass on upland areas as trial results and industry experience has shown that it will feed and fatten livestock at high stocking rates during the wet season and carry up to 40 times more cattle than native pastures. It is a grass that requires a high level of management for it to remain productive and not to become a weed in other areas. DBIRD is not recommending new sowings of gamba grass (Cameron, A. G. 2000).

References

Cameron A. (2005), NT Pastures Branch, Pers Comm. Cook, G. & Liedloff, A. (Unpublished). Flame Simulation Model. Douglas, M. M. & Setterfield, S. A. (2005). Impacts of exotic tropical grasses: lessons from gamba grass in the Northern Territory.In: Proceedings of the Eighth Queensland Weed Symposium. pp 69-73. Flores, T., Setterfield, S. A. & Douglas, M. M. (2005). Seedling recruitment of the exotic grass Andropogon gayanus Poaceae in northern Australia. Australian Journal of Botany 53: 1-7. Holmes, J., Bisa, D., Hill, A. & Crase, B. (2005). A Guide to Threatened, Near Threatened and Data Deficient Plants in the Litchfield Shire of the Northern Territory. WWF Australia, Sydney Johnston, F. H., Kavanagh, A. M., Bowman, D. M. J. S. & Scott, R. K. (2002). Exposure to Bushfire Smoke and Asthma: an Ecological Study. MJA 176: 535-538. Kerrigan, R., Cowie, I. & Baker, B. (2002b). Threatened Species of the Northern Territory: Grevillea longicuspis McGill. Christmas Holly Grevillea. Parks and Wildlife Commission of the Northern Territory, Darwin. Lemcke, B. & Cameron, A. G. (1996). Management of Kent Gamba grass: Agnote. Northern Territory Government. Lonsdale, W. M. & Lane, A. M. (1990). Tourist vehicles as a vector of weed seeds in Kakadu National Park: Report to Australian National Parks and Wildlife Service. CSIRO Division of Entomology, Tropical Ecosystems Research Centre, Darwin.