Global Invasive Species Database 100 of the worst Donations home
Standard Search Standard Search Taxonomic Search   Index Search

   Imperata cylindrica (grass)  français     
Ecology Distribution Management
and Links

      Imperata cylindrica root (Photo: Carey Minteer, University of Georgia, - Click for full size   Imperata cylindrica flower (Photo: Mark Atwater, Weed Control Unlimited, Inc., - Click for full size   Plants of Imperata cylindrica, near Nhulunbuy in Arnhem Land, Northern Territory, Australia (Photo: Colin Wilson) - Click for full size   Imperata cylindrica leaf blade portion showing off-centre midrib and scabrous leaf margins (Photo: L. M. Marsh, Florida Department of Agriculture and Consumer Services, - Click for full size   Plants of Imperata cylindrica, near Nhulunbuy in Arnhem Land, Northern Territory, Australia (Photo: Colin Wilson) - Click for full size   General appearance of Imperata cylindrica, taken on Yap, Micronesia. (Photo: Jim Space, - Click for full size   Seed head of Imperata cylindrica, near Nhulunbuy in Arnhem Land, Northern Territory, Australia (Photo: Colin Wilson) - Click for full size
    Taxonomic name: Imperata cylindrica (L.) Raeusch.
    Synonyms: Imperata arundinacea Cirillo, Lagurus cylindricus L.
    Common names: alang-alang, blady grass (English), Blutgras (German), carrizo, cogon grass (English), gi (Fijian), impérata cylindrique (French), japgrass (English), kunai, lalang, ngi (Fijian), paille de dys (French), paillotte (French), satintail (English), speargrass (English)
    Organism type: grass
    Native to Asia, cogon grass (Imperata cylindrica) is common in the humid tropics and has spread to the warmer temperate zones worldwide. Cogon grass is considered to be one of the top ten worst weeds in the world. Its extensive rhizome system, adaptation to poor soils, drought tolerance, genetic plasticity and fire adaptability make it a formidable invasive grass. Increases in cogon grass concern ecologists and conservationists because of the fact that this species displaces native plant and animal species and alters fire regimes.
    Although Imperata cylindrica can have leaf blades of up to 1.5 m tall in conditions of good soil moisture and fertility (Holm et al. 1977, in Daneshgar & Shibu 2009), the majority of its biomass occurs below ground comprising greater than 60% of the total biomass (Sajise 1976, in Daneshgar & Shibu 2009). Cogon grass is stemless erect perennial growing in loose to compact tufts with slender flat linear-lanceolate leaves arising from the rhizomes. The scabrous leaves are 4 to 10 mm wide with prominent white midribs that are slightly off center. The leaves may be 15 to 150 cm tall, depending on habitat, with narrow sharp points (Bryson & Carter 1993; Hubbard et al. 1944, in Dozier et al. 1998). For more information on morphology see Dozier et al. 1998.
    Similar Species
    Imperata brasiliensis

    Occurs in:
    agricultural areas, coastland, natural forests, planted forests, range/grasslands, riparian zones, ruderal/disturbed, scrub/shrublands, urban areas, wetlands
    Habitat description
    Imperata cylindrica occurs in a wide range of habitats, including degraded forests, grasslands, arable land, and young plantations within tropical and subtropical climates with 75 to 500 cm of annual rainfall. It can be found growing in almost all eco-types from the dryest flatwoods to the margins of permanent bodies of water. Cogon grass has invaded areas from highly xeric, upland sites to fully shaded, mesic sites. It invades sandhills, flatwoods, hardwood hammocks, sand dunes, grasslands, river margins, swamps, scrub, and wet pine savanna communities. It is known to occur from latitudes 45°N (Japan) to 45°S (New Zealand) and from sea level to over 2,000 m elevation (Langeland and Burks, 1998; Chikoye, 2003; Van Loan Meeker and Minno, 2002).

    Cogon grass typically does not invade closed forests unless they are degraded for agriculture or lumbering. It is very successful in areas that are frequently burnt, overgrazed, or intensively cultivated and rapidly colonizes such disturbed sites. A high root-rhizome to shoot ratio provides I. cylindrica a substantial source of regeneration following cutting or burning. Its rhizomes are very resistant to heat and breakage and may penetrate soil up to 1.2 m deep, but generally they only reach the top 0.15 m in heavy clay soil and the top 0.4 m in sandy soils. Its capabilies of recovery and colonization after fire allow it to take advantage of slash and burn forestry practices (Bryson & Carter, 1993, Peet et al, 1999; Chikoye, 2003; Van Loan Meeker and Minno, 2002).

    A fast-growing species, I. cylindrica thrives in areas of minimal tillage, such as orchards, lawns, and roadsides. It does not generally survive regular, deep tilling. While cogon grass is tolerant of a wide variety of soil conditions, including variations in fertility, organic matter, and moisture, it grows best in relatively acidic soils (pH 4.7). Temperature markedly affects shoot and rhizome growth, with increased growth occurring at 29º/23ºC (day/night). Temperatures of -4.5°C or lower for exposure periods of 24 hours were found to be lethal to rhizomes (Wilcut et al., 1988). While in general rhizomes do not exhibit extreme cold hardiness, stands of cogon grass may survive temperatures as low as –14ºC (Langeland and Burks, 1998; Van Loan Meeker and Minno, 2002).

    General impacts
    For a detailed account of the environmental impacts of Imperata cylindrica please read: Imperata cylindrica (cogon grass) Impacts Information. The information in this document is summarised below.

    Increases in I. cylindrica concern ecologists because this species displaces native plant and animal species and alters fire regimes (Lippincott 1997 2000, in Brewer & Cralle 2003). Dense swards of I. cylindrica create an intense competitive environment for commercially important species (Bryson and Carter 1993, Kuusipalo et al. 1995, Premalal et al. 1995, Dozier et al. 1998).

    Displacement: Invasion of longleaf pine communities by I. cylindrica will likely cause significant losses of short habitat-specialists and reduce the distinctiveness of the native flora of these threatened ecosystems (Brewer 2008).

    Agricultural: The interference of I. cylindrica with the growth of tropical crop species, both herbaceous and woody, is well documented (Brook 1989, in King & Grace 2000b).

    Habitat alteration: I. cylindrica invasion of an emerging pine forest may be an example of a grass converting a woodland with high understory diversity into a grassland with low diversity. I. cylindrica presents a case where its ability to deprive competitors of N may lead to the conversion of the ecosystem (Daneshgar & Shibu 2009).

    Modification of nutrient regime: The changes in nutrient cycling caused by exotic grasses can endanger young tree seedlings in a regenerating forest (Daneshgar & Shibu 2009). Because I. cylindrica allocates significant carbon below-ground, it is able to recover quickly after fire, which is why Lippincott (2000, in Daneshgar & Shibu 2009) suggested that frequent intense fires could convert a pine savanna into an I. cylindrica-dominated grassland.

    Ecosystem change: A study by Holly and colleagues (2008) supports the growing consensus that invasive plant species alter normal ecological processes and highlights a possible mechanism (alteration of microbial assemblages) by which I. cylindrica may alter an ecosystem process (decomposition).

    Competition: The results of a study by Brewer and Cralle (2003) suggest that I. cylindrica is a better competitor for phosphorus than are native pine-savanna plants, especially legumes. The competitive effects of this species on plant diversity may be of more immediate conservation concern relative to the effects of this species on fire regimes in longleaf pine ecosystems (Brewer 2008).

    Threat to endangered species: Longleaf pine savannas of the southeastern USA contain extraordinarily species-rich plant communities and are home to numerous threatened endemic plant and animal species (Walker & Peet 1983, Bridges & Orzell 1989, Brockway & Lewis 1997, in Brewer & Cralle 2003).

    Inhibits other species: The extensive rhizome network of I. cylindrica not only allows rapid regeneration of foliage, but also produces allelopathic root exudates that can inhibit germination and growth of other plants, including some pines (Hussain et al., 1994, in Ramsey et al. 2003).

    Modification to fire regime: Lippincott (2000) found that sandhill invaded by I. cylindrica had significantly greater fine-fuel loads that resulted in fires that had higher maximum temperatures at greater heights. Fire-induced mortality of juvenile longleaf pine was higher for pines growing in invaded sandhill.

    Other: The density of the below-ground rhizome network makes I. cylindrica a mechanical hindrance to root growth of native species. The rhizome tips are sharp: they may even penetrate the roots of native species, leading to damage or mortality by infection (Eussen & Soerjani 1975, in Daneshgar et al. 2008).

    Imperata cylindrica is used as thatch, short-term forage production, soil stabilisation and paper making (Watson & Dallwitz 1992, in Dozier et al. 1998). In surveys conducted in the coastal/derived savanna (Benin and Nigeria) and southern Guinea savanna (Ivory Coast) in Africa some farmers indicated that I. cylindrica was an important source of cheap roofing material, animal fodder and medicines. Silica bodies in the leaves contribute to its unpalatability to grazers (Coile & Shilling 1993, in Dozier et al. 1998). I. cylindrica was imported and distributed by the United States Department of Agriculture for use as a forage grass and for soil erosion control. In the United States, an ornamental variety of the grass is promoted for landscape use (Johnson & Shilling 2009). Although the ornamental varieties, known as 'Rubra,' 'Red Baron,' or 'Japanese Blood Grass' are not aggressive, plants grown from callus tissue can revert to the invasive form (Greenlee 1992, in Dozier et al. 1998).
    Geographical range
    Native range: Imperata cylindrica (L.) Beauv. var. major is native to Southeast Asia, Australia, China, Japan, the Philippines, and East Africa (MacDonald & Chandler 1994, in King & Grace 2000b). The center of origin of I. cylindrica is unknown for certain but is believed to be East Africa (Evans 1987 1991, in Van Loan Meeker & Minno 2002).
    Known introduced range: I. cylindrica is common in the humid tropics and has spread to the warmer temperate zones worldwide (Hubbard et al. 1944, in Langeland & Burks 1998). It is widely distributed in Africa, Australia, southern Asia, and the Pacific Islands, and less so in southern Europe, the Mediterranean, the Middle East, Argentina, Chile, Colombia, the Caribbean and the southeastern United States (Van Loan Meeker & Minno 2002).
    Introduction pathways to new locations
    Agriculture: Cattle ranchers wanting to improve forage spread the grass throughout Florida (Dickens 1976, in Dozier et al. 1998). It was intentionally introduced from the Philippines into Mississippi as forage (Demers et al. 2008).
    Nursery trade: A special concern is that several nurseries are now selling a cultivar of cogongrass. 'Japanese Blood Grass' or 'Red Baron', in the United States (Bryson & Carter 1993). If additional populations are found outside of the existing range, please contact the U.S. Department of Animal and Plant Health Inspection Service or the appropriate state agency.
    Other: Vegetative reproduction from rhizome segments is a significant factor in human spread of the species because these are often found in dirt moved as fill (Ayeni and Duke 1985, Willard 1988, Shilling et al. 1997, in Van Loan Meeker and Minno 2002).
    Road vehicles (long distance): Moved with heavy equipment from job to job.
    Seafreight (container/bulk): Imperarta cylindrica was accidentally introduced as packing material from Japan in Mobile, Alabama in 1911 (Demers et al. 2008).

    Local dispersal methods
    Natural dispersal (local): The aggressive nature of Imperata cylindrica is attributed to its rhizomes. These are normally concentrated in the upper 15 to 20 cm of soil (Ivens 1980 in Chikoye 2003). Rhizomes readily sprout into new shoots after fragmentation by tillage or any other form of disturbance and are resistant to fire damage (Chikoye 2003).
    On animals (local): Disperses over long distances into a variety of habitats by windborne seeds (Bryson and Carter 1993, in Langeland and Burks 1998); distances dispersed range from an average of 15m to 100 m (Holm et al. 1977, Shilling et al. 1997, in Van Loan Meeker and Minno 2002).
    On animals (local): Seed only up to several hundred metres.
    Road vehicles:
    Translocation of machinery/equipment (local): The transport of rhizomes and seeds on equipment or in contaminated soil are considered important modes of spread (MacDonald 2004, in Yager & Smith 2009).
    Wind dispersed: Plumed seeds are capable of traveling long distances in the wind, with an average flight path of 15 to 100 m (Holm et al. 1977, Shilling et al. 1997, in Van Loan Meeker & Minno 2002). The hairy seeds tend to catch on nearby vegetation.
    Management information
    For more detailed information on management of Imperata cylindrica (cogon grass) please read: Imperata cylindrica (cogon grass) Management Information. The information in this document is summarised below.

    Preventative measures: A Risk assessment of Imperata cylindrica for the Pacific region was prepared by Pacific Island Ecosystems at Risk (PIER) using the Australian risk assessment system (Pheloung 1995). The result is a score of 19 and a recommendation of: reject the plant for import (Australia) or species likely to be a pest (Pacific).

    Results from a study by King and Grace (2000a) suggest that efforts to prevent I. cylindrica invasion should focus on preventing I. cylindrica propagules (seeds and rhizomes) from reaching sensitive communities.

    Chemical control: Ramsey and colleagues (2003) report that herbicides may temporarily control I. cylindrica foliage up to 12 to 24 months (Willard et al. 1996 1997, in Ramsey et al. 2003). Without the re-establishment of desirable species, viable rhizomes will eventually allow cogon grass to re-infest the area (Shilling et al. 1995, in Ramsey et al. 2003). Controlling rhizomes with herbicides is a difficult task. A combination of glyphosate applied at 2.8 kg ai ha_1, followed by fertilization and reseeding with Bermuda grass (Cynodou doctylon) showed that rhizome reserves were sufficient for cogon grass to recover from both treatments (Johnson 1999, in Ramsey et al. 2003). In another study, cogon grass recovered after imazapyr was applied at 2.24 kg ai ha_1 in a 2-year-old loblolly pine plantation (Miller 2000, in Ramsey et al. 2003). Even these relatively high herbicide application rates there is still a remnant population of viable rhizomes that has the potential to re-infest the treated site. Further research is needed to integrate herbicide usage, which provides short-term control and a ‘‘window for re-establishment’’, with bio-control using desirable, yet highly productive plant species for long-term control (Ramsey et al. 2003).

    Recommendations (Demers et al. 2008) to control cogon grass in the southeastern USA are to treat infestations in autumn (May through to October) with glyphosate and/or imazapyr herbicides. These recommendations are consistent with a wide range of studies conducted (Miller 2003, Faircloth et al. 2005, in Demers et al. 2008). Fluazifop is also an effective option (Demers et al. 2008). Guidelines for herbicide control are detailed by Demers and colleagues (2008).

    Integrated management : Shade, repeated herbicide application, and mechanical control have all been used to control I. cylindrica (Macdicken et al. 1997, Terry et al. 1997, in Brewer & Cralle 2003). An integrated approach that combines burning, tillage (mechanical control) and chemical applications provide the best approach for cogon grass management (MacDonald et al. 2009). Cogon grass should first be burned or mowed, preferably in summer, to remove excess thatch and older leaves. Subsequent regrowth (of one to four months) will reduce rhizome biomass and allow herbicides to target actively growing leaves which maximises herbicide effectiveness. Once control of cogon grass has been achieved planting of desirable vegetation should occur as quickly as possible to prevent reinvasion (MacDonald et al. 2009).

    Other: Brewer and Cralle (2003) found that short-lived, high-level pulses of phosphorus addition reduce the competitive advantage I. cylindrica has over native plants without negatively affecting native plant diversity. Additional work is needed, however, to elucidate the mechanism of inhibition of I. cylindrica by P addition (Brewer & Cralle 2003).

    One of the reasons for the high invasive potential of Imperata cylindrica is its ability to grow on a wide range of soil types with no special nutrient requirements. It has extremely efficient nutrient uptake and forms associations with mycorrhizae which enhance nutrient uptake on unfertile soils (Collins & Jose, 2009 and references therein).

    I. cylindria assimilates carbon via the C4 photosynthetic pathway (Paul & Elmore, 1984) giving it a competitive advantage over C3 plants in many conditions and contributing to its invasiveness. It is a strong competitior for water, light and nutrients because it sprouts and grows more rapidly than most crops. However, like other C4 plants it is relatively intolerant of shade.

    The ecological resiliency of Imperata cylindrica and its ability to regenerate from any man-made or natural disturbance, is primarily attributed to the well-protected rhizome network. The rhizome biomass can accumulate up to 1100 g m-2 in a mature sward (Soerjani 1970, in Ramsey et al. 2003). I. cylindrica can reproduce asexually from rhizome fragments as small as 0.1 g (Ayeni & Duke 1985, in Daneshgar & Shibu 2009). It is a prolific seeder, producing as many as 3 000 seeds per plant (Holm et al. 1977, Daneshgar & Shibu 2009). Indications are that seed viability is extremely short-lived.
    Lifecycle stages
    Rhizome production from a seedling plant takes about 4 wk. Seed viability is short lived with no dormacy. As many as 4.5 million shoots have been reported more than 10 metric tons of leaf material and more than 6 metric tons of rhizomes produced from a single hectare (Bryson & Carter, 1993).
    This species has been nominated as among 100 of the "World's Worst" invaders
    Reviewed by: Dr. James Leary, Department of Natural Resources and Environmental Management, College of Tropical Agriculture and Human Resources, University of Hawaii at Manoa
    Compiled by: Profile revision: National Biological Information Infrastructure (NBII) & IUCN/SSC Invasive Species Specialist Group (ISSG)
    Last Modified: Saturday, 21 August 2010

ISSG Landcare Research NBII IUCN University of Auckland