Cookies on Plantwise Knowledge Bank

Like most websites we use cookies. This is to ensure that we give you the best experience possible.

Continuing to use means you agree to our use of cookies. If you would like to, you can learn more about the cookies we use.

Plantwise Knowledge Bank
  • Knowledge Bank home
  • Change location
Plantwise Technical Factsheet

common reed (Phragmites australis)

Host plants / species affected
Beta vulgaris var. saccharifera (sugarbeet)
Brassica napus var. napus (rape)
Gossypium (cotton)
Helianthus annuus (sunflower)
Oryza sativa (rice)
Saccharum officinarum (sugarcane)
Solanum tuberosum (potato)
Triticum aestivum (wheat)
Zea mays (maize)

Typical P. australis in Eurasia is a robust erect perennial grass, aquatic or subaquatic, growing to 4 m high (occasionally 6 m), strongly tufted, with an extensive rhizome system. Stolons may also be present. Stems rigid, many-noded; internodes hollow. Leaves alternate, up to 70 cm long, with a ligule of hairs (resembling short eyelashes) up to 1.5 mm long; leaf blade flat, up to 60 cm long and 8-60 mm wide, tapering to a spiny point, rigid, glabrous or sometimes covered with a whitish bloom; leaf sheaths loose and overlapping. Inflorescence a feathery, drooping panicle 15-50 cm long, often tan-brown to purplish; many-flowered; branches slender, ascending; spikelets several-flowered, 10-18 mm long, with florets exceeded by rachilla hairs; first glume 2.5-5 mm long; second glume 5.7 mm long; lemmas thin, 3-nerved, densely and softly hairy; nerves ending in slender teeth, the middle tooth extending into a straight awn; seed slender, dark brown.

The native North American P. australis subsp. americanus is distinguished by basal internodes red or purplish (pale yellow in typical  introduced forms) and lower glumes up to 7 mm long (up to 5 mm in introduced forms) (Catling, 2006).Other differences include ligule width 1-1.7 mm in subsp. americanus; 0.4-0.9 mm in exotic forms; and leaf sheaths loose and shed as the plant senesces in subsp. americanus, tight and retained after the leaf falls in exotic forms (Saltonstall, 2005).

Prevention and control

Effective management programmes for Phragmites are usually based on burning, the use of plastic mulch, discing, chemical control, cutting, grazing, dredging, and draining or the manipulation of the water table and salinity (Marks et al., 1994).

Cultural Control

Cutting is a standard control measure, but tends to produce highly variable results. Nearly two years effective control was achieved in North Carolina marshes, USA, using a hand-held weed trimmer with a saw blade to clear vegetation to within 15 cm of the hydrosoil surface Kay (1995). However, in UK drainage channel systems, frequent winter mowing promoted growth of P. australis and adversely affected drainage efficiency (Milsom et al., 1994). Mowing every 2, 4 or 8 weeks achieved 93, 81 and 69% reduction at the end of the growing season, respectively, but there was strong re-growth in the following season (Derr, 2008). In Japan, Asaeda et al. (2006) showed cutting in June to be better than cutting in July. Smith (2005) showed that on a small scale, repeated manual breakage of stems below water could achieve long-term control.

Rolletschek et al. (2000) confirm that cutting and burning can be much more effective when followed by flooding. Cutting or other harvesting methods are often considered to be the least-damaging of available control methods on the environment (Ditlhogo et al., 1992), and there were no consistent deleterious effects on herbivorous arthropods of regular harvesting of reeds in Lake Neusiedler, Austria (Kampichler et al., 1994).

Autumn ploughing 25-27 cm deep for rhizome exposure and desiccation followed by reploughing in spring is effective against the common reed in Russian rice crops. Fragmented rhizomes soon die under a layer of water and this method is reported to be very effective if carried out every year (Agarkov, 1980).

In Dutch reed stands, Mook and Toorn (1982) found that damage to just-emerging shoots in April or early May by burning (wet- and dry-burned plots) or early ground-frost (dry-burned and dry-mown plots), retarded growth of leaves and shoots for about 1-2 weeks, but the relative growth rate and maximal levels reached were not significantly lower than in undamaged areas. Heavy damage by late ground-frost (dry-burned plot) or by the stem-boring larvae of Archanara geminipuncta (wet-undisturbed plot) lowered the maximal shoot biomass by about 25-35%. Heavy infestation by the rhizome-boring larvae of Rhizedra lutosa (dry-undisturbed and dry-mown plots during later years) gave losses in yield of about 45-60%.

Winter harvesting and burning influence the geometry (stem diameter and relative wall thickness) and mechanical properties (modulus of elasticity and breaking stress) of reed culms (Ostendorp, 1995), weakening the plants and making them more susceptible to other control measures.

Biological Control

Schwarzländer and Häfliger (2000) record 28 insect herbivore species feeding on P. australis in North America, and more than 140 insect species in Europe and Asia Minor. For at least 55 of these, P. australis is the only known host plant. A range of shoot flies, gall midges (Cecidomyiidae), and moths cause damage by mining in stems or rhizomes and are considered to have some ‘minor potential’ as biocontrol organisms but none have been fully checked and tested. Häfliger et al. (2006a,b) consider Archanara geminipuncta to be the most promising organism to be studied so far for control of invasive populations in North America, but there have been no reports of its implementation. Some attention has been paid to the possibilities of inundative control of Phragmites reeds using insects in Australia, although the approach does not appear to have been widely used there (Wapshere, 1990).

Grass carp (Ctenopharyngodon idella) will eat the young shoots of developing Phragmites reeds, but they are not generally a preferred species (Nikanorov and Polyakova, 1980). However, in fallow rice paddies in Japan, Tsuchiya (1979) reported that grass carp weighing 0.5-2 kg grazed areas of 4-30 m² in fallow paddies, controlling P. australis, as well as Typha latifolia and Isachne globosa.

Chemical Control

Glyphosate is usually highly effective against Phragmites and other emergent weeds, across a wide spectrum of climatic and environmental conditions (Barrett, 1976; Riemer, 1976; Comes et al., 1981; Eaton et al., 1981; Fernandes et al., 1981; Evans, 1978,1982; Baird et al., 1983; Al-Juboory and Ali, 1996). However, Derr (2008) cautions that there is almost invariably some regrowth after 12 months and that repeat applications are required for continued control. Glyphosate has been successfully used to control invasive reed growth in sensitive areas, such as wetland bird reserves in Spain and the UK (Cooke, 1991; Dies Jambrino and Fernández-Anero, 1997), and freshwater fisheries (Caffrey, 1996). Arsenovic and Konstantinovic (1990) also found imazapyr was 100% effective against emergent weeds including P. australis whereas glufosinate application resulted in 98% weed control and 10-20% regrowth. However, Derr (2008) reports only partial control by glufosinate. Dalapon selectively controls Phragmites and other narrow-leaved emergent monocotyledonous weeds (Chancellor, 1960; Barrett, 1976; Agaronian et al., 1980; Comes et al., 1981), being particularly effective when sprayed at the time of year, usually mid-late summer, when carbohydrate reserves are being laid down in the reed rhizome (Barrett and Robson, 1974).

Weed-wiper technology has been employed using glyphosate and imazapyr against Phragmites reeds, but with mixed success (Evans, 1982). In Sweden, Svensson (1983) reported 80% control of P. australis in spring wheat and potatoes, but the technique was not successful against reeds in waterways. Kay (1995) concluded that whilst cutting can provide long-term reed control, wipe-on herbicide application using reduced-dose treatments comprising 25-50% of the recommended concentration was not a practical method of P. australis control in a shallow, freshwater marsh in North Carolina, USA. In Italy, Rapparini and Fabbri (1988) reported that contamination of the water in canals treated with glyphosate to control reeds was reduced to a minimum or even eliminated when the herbicide was applied with a 4-m smear boom fitted with brushes, and both banks and channels were able to be treated in this way.

Aerial application of glyphosate and imazapyr using a microlight aircraft was successful in controlling Phragmites stands causing a fire hazard along power line routes in South Africa, giving 50-100% control, with cost savings of 72% compared with conventional land-based control operations (Rensburg, 1996).

Flupropanate has been used for reed control in uncultivated and forest areas in Japan and Korea (Manabe, 1980), although symptoms of yellowing, twisting, stunting and necrosis were observed in crops screened by Hwang et al. (1996). Both direct-seeded and transplanted rice were also sensitive to the herbicide. Effects on reeds varied with soil type of the treated field and reed growth stage and a relatively high dosage of the herbicide was required for reed control. In Japan, haloxyfop provided excellent long term control of P. australis when the weed was 1-1.5 m tall (Matsumoto, 1987). In laboratory studies, Xian and Price (1987) found that fluazifop-butyl could give effective control and suppression of regrowth of P. australis and was the most effective herbicide for post-emergence control of reed in soyabeans and cotton. Derr (2008) reports negligible effectiveness of a range of grass herbicides including fenoxaprop, fluazifop, clethodim, sethoxydim, MSMA, quinclorac and dithiopyr.

In flax crops in Romania, successful attempts were made to control P. australis using bromoxynil + MCPA (post-emergence), together with either a pre-sowing treatment with metolachlor or with the addition of fluazifop-butyl to the bromoxynil + MCPA treatment (Sarpe et al., 1981).


Integrated control may be used in eradication programmes aimed at Phragmites reeds. In Egypt, Ashour Ahmed (1990) reported that a regime consisting of 2 applications of dalapon combined with burning 10 weeks after the second application, followed by ploughing in conjunction with the removal of dead rhizomes with the ploughed soil, and finally flooding, gave near-complete control of a dense stand of P. australis in a dry pond of 3000 m². Less than 5% of the initial plant cover had regenerated 10 months after herbicide application.

Cane growers in southern Queensland, Australia, have advocated an integrated programme for reed control using deep ploughing followed by rotary hoeing, which can destroy most of the root system of P. australis, but this is unlikely to provide complete eradication. Two applications of dalapon gave rapid, short-term control, whereas two applications of glyphosate showed slower, but longer-term control (Izatt, 1979).

Summary of invasiveness

Phragmites australis, the common reed, is an aggressive, vigorous species which, in suitable habitats, will out-compete virtually all other species and form a totally dominant stand. Its invasive character has been particularly apparent in North America where it has become dominant in a range of wetland habitats replacing native species and biotypes including the native North American P. australis subsp. americanus. Bird, fish and insect populations can also be affected.

Related treatment support
External factsheets
IRRI Factsheets, International Rice Research Institute (IRRI), English language
Zoomed image