Alternatives

Not to be confused with There is no alternative.

Alternatives to giant hogweed are the plants and uses of land recommended in its place: ornamental plants to grow instead of it in gardens, grasses and other crops to sow on land from which it has been removed, and grazing animals to keep that land clear once control has ended.¹² Lists of garden substitutes have been published in Britain, Canada, Belgium and Switzerland, and the European Giant Hogweed Best Practice Manual of 2005 advises that cleared land be sown densely with grass or grazed.²³⁴⁵¹citation supplied

The sources also give costs. In Latvia, at 2005 prices, a first year of control that included sowing grass cost four to six and a half times as much as two treatments with glyphosate.⁶ Of 475 public contracts for destroying the plant in Russia in 2024, one, for 0.8 ha, included sowing grass; 289 paid for herbicide alone.⁷ Grazing is cheap where land can be fenced, but slow: estimates run from two years to reduce a stand to ten or more to eliminate it.¹⁸⁹

The codes that ask the plant trade to offer substitutes are voluntary; where the law requires control, as in Latvia, Ontario and Russia, the duty falls on those who hold the land.^1011121314 A European research project, Giant Alien, looked for a natural enemy of the plant between 2002 and 2005, and found none.¹⁵

In gardens

Giant hogweed was grown for its size. In 1870 William Robinson recommended the “Giant Cow Parsnip, Heracleum giganteum” for rich soil near river banks, or wherever a striking effect was wanted at a distance; in 1883 he added that such plants sow themselves and should not be allowed to monopolise the ground.¹⁶¹⁷ The guides that now offer substitutes describe what gardeners wanted in the same terms: stature and large leaves, or “bold, tropical-looking leaves”, huge white umbels and a “sturdy, architectural look”.³¹⁸ A British booklet of 2022 recalls the plant’s introduction to Victorian gardens as a “monumental curiosity”, and files its replacements under “Architectural plants”.²

Britain and Ireland

In 2010 the wild-plant charity Plantlife and the Royal Horticultural Society (RHS) published Gardening without harmful invasive plants. About 60 per cent of the invasive non-native plants damaging British habitats, it said, had come from gardens and garden ponds, and many hundreds of thousands of pounds a year were spent on limiting their spread; the government had powers to ban their sale, but had banned none.¹⁹ The booklet did not mention giant hogweed. Its “safer tall feature plants”, among them purple angelica (Angelica gigas), ornamental rhubarb and a rodgersia, were offered in place of Japanese knotweed.¹⁹

Its successor of 2022, issued under the name Be Plant Wise by the GB Non-native Species Secretariat and the Animal and Plant Health Agency and paid for by nine water companies, lists over 160 plants, revised in 2021 from the list of 2010 and chosen for a low risk of invasion. It speaks of “millions of pounds” spent each year.²²⁰ It pairs giant hogweed with eighteen alternatives, from grasses and bulbs to angelica-tree (Aralia cachemirica), Cretan hemp, ornamental rhubarb, melancholy thistle, lady fern, an iris called ‘Black Swan’ and a dark-leaved cow parsley, Anthriscus sylvestris ‘Ravenswing’; the descriptions stress size, with spikes of umbels and stems up to 3 m tall.² In its table of characteristics three of the eighteen carry the mark “RHS Plants for Pollinators”, two are marked as UK natives and seven as giving shelter or cover to wildlife.²chosen for the architecture Purple angelica, which the agency’s blog presented as a plant for pollinators, is in the booklet too, but in place of Himalayan balsam.²²⁰ A companion booklet for landscapers offers sixteen plants in place of giant hogweed.²¹

The RHS gives no substitutes on its own advice page. Having the plant in a garden is not an offence, it explains, provided it is kept from spreading; it advises removal, and calls hiring a professional “the easiest and safest way” to do it.²² Invasive plants, it notes elsewhere, are expensive to eradicate, and restoring degraded habitat is very costly, if it can be done at all.²³

In Ireland a voluntary code for horticulture, first published by Invasive Species Ireland in 2008, says in its draft revision of 2012 that substitutes should be suggested and offered to customers, which lets the nursery trade present an innovative and “green image”, in the same words as the European code; it names none for giant hogweed.²⁴¹¹ Irish guidelines on the plant say that after an overall spray with glyphosate it may be necessary to reseed with grass and other native plants, since a good sward soon after treatment slows recolonisation from seed.²⁵

Canada

The Ontario Invasive Plant Council’s Grow Me Instead guide for Northern Ontario (2014) offers a wild angelica, which it names “Angelica purpurea (syn. A. stricta)” and whose sap, it warns, may cause allergic reactions; ornamental rhubarb, for its leaves 60–90 cm long; and rodgersias.³ Giant hogweed has not been sold in Ontario since it was listed as a noxious weed in 2010, the guide notes, but gardeners may still pass it on through plant or seed swaps.³ By 2020 the programme had distributed 73,000 guides and 6,350 packets of native seed.²⁶

The guide for British Columbia (2013) lists blue elderberry (Sambucus cerulea), a native shrub or small tree 5–10 m tall with flat-topped clusters of white flowers; Ligularia dentata; rodgersias and shieldleaf rodgersia (Astilboides tabularis); and the four species of Angelica native to the province.¹⁸ A landscaping company in Ontario has recommended the native cow parsnip, Heracleum maximum, as similar in appearance “but without the hazardous sap”;²⁷ Ontario’s guide to managing giant hogweed says that exposure to cow parsnip has similar effects.¹³

United States

In the United States, where giant hogweed is a federal noxious weed, the Morton Arboretum in Illinois recommends common boneset (Eutrochium perfoliatum) in its place, for its similar inflorescence and habit, and the Johnson Creek Watershed Council in Oregon names Pacific ninebark (Physocarpus capitatus), a native shrub with white flowers.²⁸²⁹³⁰ New York advises replanting with native or non-invasive plants after removal, and Maine establishing replacement vegetation as soon as possible after treatment with herbicide.³¹³²³³ The regional guides to non-invasive garden plants for western Washington and for Oregon do not include giant hogweed.³⁴³⁵

Continental Europe

In Belgium the LIFE project AlterIAS (2010–2013, budget €1,010,804) noted that native look-alikes of invasive plants are rarely available, and proposed natives with a similar function instead.³⁶⁴ Its brochure of 2012 pairs giant hogweed, as a plant of the flower border, with valerian, which it describes as attractive to pollinators and names Valeriana officinalis in one place and V. repens in another; for such borders it also offers columbine, monkshood, foxglove, great mullein and other natives.⁴

A Swiss list of 2007, drawn up by Andreas Gigon of ETH Zürich for the Swiss Commission for the Conservation of Wild Plants, proposes the native Angelica sylvestris, Peucedanum verticillare and Molopospermum peloponnesiacum and the non-native Angelica archangelica and Rheum palmatum, chosen to resemble giant hogweed as closely as possible in use, size, form, leaves, colour and flowering time. Whether they could be bought was not a criterion.⁵

The code of conduct on horticulture of the Council of Europe and the European and Mediterranean Plant Protection Organization (EPPO) includes the practice “Make substitutes for invasives available”, which, it says, also gives the trade “an innovative and green image”. It warns that substitutes may themselves turn invasive and may not suit another country, and its worked examples include no hogweed.¹¹

After control

See also: Control of giant hogweed § After control

Land cleared of giant hogweed is left bare and open to erosion and reinvasion. The Best Practice Manual recommends returning it to regular use, above all cropping or the management of grassland, with a cover crop to protect the soil.¹ The Giant Alien project’s models concluded that control must last about five years or longer, at a high intensity, or the populations recover.¹⁵

The Best Practice Manual

The manual’s advice comes from experiments on former silage fields of Sosnowsky’s hogweed, overrun after cultivation was abandoned and grazing animals declined. On such land it recommends glyphosate in spring and deep ploughing three weeks later, which almost totally stops germination, and then grasses sown at a high density, 4,000 emerging seedlings per square metre: competitive, dense, cut-tolerant native species, such as Dactylis glomerata with Festuca rubra, or Lolium perenne with F. rubra and Poa pratensis.¹ On riversides no herbicide is used: the plants are cut in spring, grass is sown by hand at a higher rate, and the sward is cut often. On sandy floodplains, where floods leave hogweed seed in the sward, the manual advises against sowing at all.¹

Grasses that establish themselves, if cut often, also form a competitive sward, which grows more diverse as native broad-leaved plants arrive; the land can then be farmed or used for recreation. Trees are a special kind of cover crop: beech shades out tall hogweeds well, firs, alders and willows less so.¹ The manual prefers an integrated strategy, judged on “efficacy, ecology and economy”, to any single method, and says that the sowing and planting of tall invasive hogweeds in gardens, parks and the open landscape “should be prohibited”.¹

Latvia

The method was tried in Latvia within the Giant Alien project. On a farmer’s field in Gauja National Park, uncultivated for five years and carrying 25–30 plants of Sosnowsky’s hogweed per square metre, plots were sown on 13 May 2003 at 4,000 germinated seeds per square metre with the manual’s mixtures.³⁷ Most of the hogweed came from seed rather than from surviving roots. In the trial of 2004, with glyphosate on 6 May, ploughing to 24 cm on 30 May and sowing on 1 June, no Sosnowsky’s hogweed was recorded. Herbicide in the young sward reduced its diversity; cutting did not.³⁷

A Latvian government booklet of 2007 repeats the manual’s mixtures and rate.⁶ On land reclaimed in Barkava parish, ploughing and a first-year crop of annual ryegrass, oilseed radish or buckwheat left 5–10 per cent of the hogweed plants, and barley and spring wheat followed. The booklet recommends green manure and then green-fodder or silage crops cut in June, and warns that unless the land goes on being managed in later years, the money spent on control is wasted.⁶

Québec

In an experiment at the Montréal Botanical Garden, hogweed seed was spread on 48 plots sown with five seed mixtures or left unseeded. Germination was highest on the unseeded plots; a commercial mixture dominated by grasses gave the lowest, and no hogweed seedling survived in it. The more cover, the less hogweed germinated, survived and grew.³⁸³⁹ A mixture rich in Eupatorium was included because the researcher had seen fewer hogweeds where boneset grew.³⁸ The thesis concluded that a mainly grass mixture that quickly forms a dense cover has good potential for restoration.³⁸ A Swedish review of 2025, reading the study as showing Canada goldenrod to be particularly effective, noted the irony that the plant is itself regarded as invasive in Sweden.⁴⁰

Russia

A review of 35 Russian sets of recommendations from 2004 to 2024 found chemical treatment recommended most often, followed by mowing, in 29, and agronomic measures, in 26, which were said to bring abandoned farmland back into use. Half of the recommendations also included dubious methods, such as burning, salting the roots or petroleum products.⁷ The contracts show what was paid for. Of 475 public contracts in 2024, digging, agronomic measures and sowing grass appeared in 3 per cent. A single contract, for 0.8 ha at 1.772 million roubles, combined mowing, ploughing and sowing grass; 289, for 21,553 ha at 242.6 million roubles, paid for herbicide alone.⁷289 to spray, one to sow In all, about 42,000 ha in 33 regions were treated for about 650 million roubles, and mowing, at 24,800 roubles a hectare, cost 1.7 times as much as herbicide.⁷

The Kurgalsky nature reserve in Leningrad Oblast, through whose southern part the Nord Stream 2 gas pipeline runs, was cleared by digging. Nord Stream 2 AG commissioned a study of how the hogweed had entered the reserve, and paid for three years of removal by hired workers.⁴¹the pipeline paid for the weeding Two meadow plots were then cut by machine and sown with Festuca pratensis, Bromopsis inermis and Poa pratensis. The authors report that the plant was eliminated from the reserve.⁴¹

Several Russian methods are patented, among them sowing grass through a geotextile laid under the soil, and mowing twice, spraying, and sowing grasses and red clover three weeks later.⁴²⁴¹ A review of such methods found mowing alone only 5.2–12.1 per cent effective, and called a dense turf of sown grasses an ecologically clean and fairly effective method.⁴² Another review proposes turf grasses sown at 150 kg a hectare, Jerusalem artichoke, itself invasive, and goat’s rue (Galega orientalis) as competitors, and spruce or Scots pine on waste ground.⁴³

In the first season of a field trial in Moscow Oblast in 2024, deep ploughing killed 78 per cent of the plants, mowing 50 per cent and a glyphosate product 97 per cent. The competitors sown, among them lupin, lucerne and Galega orientalis, lacked light under the hogweed; only Jerusalem artichoke grew regardless. The authors judged herbicides the most expensive method in direct costs and tillage the cheapest, and planned to add milk thistle and Sakhalin knotweed in 2025.⁴⁴⁴⁵

Poland

Polish official guidelines of 2014, written by the foundation “Palący Problem – Heracleum” for the General Directorate for Environmental Protection, recommend revegetation at the manual’s density and with its mixtures, since clearing a dense stand often destroys the whole plant cover; any strategy, they say, should include it.⁸ Their model tender ends with a phase of “grace” (faza karencyjna), in which competitive native vegetation is re-established and the land kept in use, for example by permanent grazing.⁸

Cost

In Latvia, at 2005 prices, the first year of control of a medium-dense stand cost 38–100 lats a hectare for mowing with a tractor, 47–62 for two treatments with glyphosate and 87–126 for soil cultivation; the integrated method cost 247–312 lats, of which sowing grass accounted for 120–150 (Fig. 1). It thus cost about four to six and a half times as much as chemical control, and the sowing alone about two to three times as much.⁶

████▒▒▒▒▒▒                          38–100
█████▒                               47–62
█████████▒▒▒▒                       87–126
· · · · · · · · · · · · · · · · ·
████████████▒▒▒                    120–150
█████████████████████████▒▒▒▒▒▒    247–312
+---------+---------+---------+--
0        100       200       300

A German cost–benefit study included restoration in the price of chemical control, with a seed mixture of about 70 per cent grass and 30 per cent herbs sown at 4,000 seeds or 20 g per square metre. It put the minimum cost of eradicating the plant in Germany over ten years at €3.47 million, or €6.25 million with two reinfestations, against benefits, estimated from what people said they would pay, of €238 million a year, or €59.5 million after correcting for overstatement.⁹

Grazing

See also: Control of giant hogweed § Grazing

Grazing can outlast control as a use of the land. The Polish guidelines end their model tender with permanent grazing, and the Latvian booklet calls fodder crops or grazing the most profitable use of infested land during control.⁸⁶ According to the Best Practice Manual, grazing is cheap where large areas can be fenced, and over time it creates a dense sward of grazing-tolerant species that leaves little ground for seedlings. Livestock need time to get used to hogweed, but then prefer it.¹ The manual’s example from Denmark, 111,800 plants on 9.18 ha along the stream Seest Mølleå, priced steel fencing for sheep, at 2002 prices, at €2,107 a year, not counting the animals or the labour, against first-year costs of €20,196 for two treatments with herbicide, not counting the chemicals, and €36,894 for root cutting. For a large colony it estimates 1,000 hours a year to inspect and move 170 sheep across ten areas.¹

In Scotland the Scottish Invasive Species Initiative grazed sheep from 2019 to 2022 on 7 ha of riverside woodland near Macduff, on the River Deveron. The first flock, 25 sheep, overgrazed the site. The sheep-days were then cut from 5,076 in 2019 to 1,326 in 2021, and the grazing pressure from 0.3 to 0.07 livestock units per hectare a year by 2022, without loss of control; in 2021 the number of hogweed plants fell by 95 per cent between April and September.⁴⁶⁴⁷ The initiative advises grazing from early May to mid-August, at low pressure, and still cutting any flower heads in July.⁴⁷ Herbicide was, to the initiative, “the alternative control method”: unlike glyphosate, grazing leaves the ground flora in place, as a barrier to new seedlings.⁴⁸the flock keeps what it clears

In Poland the only pasture in the invaded village of Górki had no flowering hogweed, according to unpublished data of the foundation that wrote the national guidelines.⁸ In Latvia, beef cattle on a farm in Zaube parish ate the plant and kept it from flowering, although another source held that no animal eats it fresh.⁶ A Russian review counts grazing by cows, goats and sheep as environmentally harmless, but reports lesions on the udders and heads of dairy cattle on infested pasture, and pinkish milk from cows with damaged udders.⁴³ No published study, according to the Swedish review, has tested cattle, goats or pigs as a method of control.⁴⁰

Estimates of the time needed differ: two years to reduce a stand and five to eliminate it, according to EPPO as cited by the Polish guidelines; usually five to ten years, according to Buttenschøn and Nielsen as cited in the Swedish review, which also cites a Danish site where seven years of sheep grazing exhausted the seed bank; and at least ten years, according to the German cost–benefit study and a Ukrainian review.⁸⁴⁰⁹⁴⁹ In Maine, grazing is said to help, but usually to suppress the plant rather than destroy it.³³

For bees and for fodder

Beekeepers valued giant hogweed for its many flowers and often spread its seed; their interest waned after warnings about the plant from the middle of the twentieth century.⁵⁰ In Germany they probably sowed it into the wild.⁵¹ A Belgian brochure of 2012 still describes it as a nectar plant highly appreciated in beekeeping.⁴ In Poland in the 1970s and 1980s Sosnowsky’s hogweed was promoted as a high-yielding fodder plant and a new source of nectar, and its seed was circulated among beekeepers and sent by post.⁸ A Russian paper of 2024 credits it with up to 300 kg of honey a hectare “by literature data”, without citing a primary study.⁴⁵

The evidence on bees points both ways. At 34 sites in southern Poland, honeybees were up to four times as abundant in severely invaded plots as in uninvaded ones, at the expense of other pollinators.⁵² A study from Pskov Oblast lists hayfields overgrown with the plant, together with the decline of lime trees and the loss of flowering herbs to herbicides, among the causes of a poorer forage base for bees, without measuring its share.⁵³ The lists of substitutes are written for gardeners, not beekeepers: the Belgian substitute, valerian, is described as attractive to pollinators, and three of the eighteen British alternatives carry the RHS mark for pollinators, while the crops proposed in Russia to compete with the hogweed are proposed as competitors, not as forage for bees.⁴²⁴⁵

Sosnowsky’s hogweed itself was grown for silage, and was given up, according to a Russian review, because of the bitter taste and smell it gave to milk and meat and the harm it did to the animals’ reproduction;⁴³ in Estonia it was still being propagated as a good fodder plant in the late 1980s.⁵⁰ The first crops on reclaimed Latvian land were annual ryegrass, oilseed radish and buckwheat, then barley and spring wheat.⁶ Russian papers now propose goat’s rue, which, like the hogweed, comes from the Caucasus, together with lupin, sweet clover, lucerne, amaranth and white mustard, as crops to compete with it.⁴⁵⁴³

The Giant Alien project

Giant Alien was a project of the European Union’s Fifth Framework Programme, with the full title “Giant hogweed (Heracleum mantegazzianum) a pernicious invasive weed: developing a sustainable strategy for alien invasive plant management in Europe”. It ran from 1 January 2002 to 30 April 2005 at a total cost of €2,934,773, of which the EU contributed €1,721,459, and was coordinated by the Royal Veterinary and Agricultural University at Hørsholm, in Denmark.⁵⁴ Its size is given variously: more than 40 scientists from seven countries in the manual, 37 researchers from eight institutions on its website; the manual and CORDIS also list different partners.¹⁵⁵⁵⁴ Its aim was a sustainable strategy for controlling the plant that could serve as a template for other invasive species.⁵⁵

The project produced the Best Practice Manual (2005), printed in 2,000 copies and published in eight languages, and the monograph Ecology and Management of Giant Hogweed (2007), and it listed revegetation among its exploitable results.¹¹⁵⁵⁶ It found no suitable agent for classical biological control, a conclusion its final report called “a startling one, and most unusual for a weed biological control project”, and it found biological control with pathogens “inadvertently blocked” by the wording of EU Directive 91/414 (see Enemy release hypothesis § The search for biological control).¹⁵

Codes and law

Voluntary codes

The British Horticultural Code of Practice, published in 2005 and revised in 2011, tells suppliers to “Provide substitutes for invasive plants”. Breaching it is not an offence, although a court must take it into account.¹⁰ It was drawn up by a working group of Defra, the Scottish Executive, the Welsh Assembly Government, Gardening Which?, the RHS, the National Trust, Plantlife International and Kew, with three trade associations: the Garden Centres Association, the Horticultural Trades Association and the Ornamental and Aquatic Trades Association.¹¹ The Canadian Grow Me Instead guides acknowledge the Australian Grow Me Instead programme of the Nursery and Garden Industry Australia.³

The European code of the Council of Europe and EPPO, first published in 2009, rests on self-regulation. Its reasons, it explains, are that the trade deals mainly in non-essential goods, that equally appealing non-invasive plants can often be used instead, and that “the threat of increased government regulation can act as a motivation”.¹¹the threat does the regulating It reports that the RHS had barred exhibitors from selling or showing giant hogweed at its shows since 2004, and that plant retailers in Picardy had signed a voluntary charter to withdraw it from sale within six months.¹¹ In Belgium, a voluntary code launched in 2011, whose signatories agree not to sell or plant 28 species, giant hogweed among them, had over 1,000 adherents after two years, among them an estimated 10–30 per cent of the professionals in federations. Its report compares the 56 per cent of nursery professionals who had heard of it with the 7 per cent of American nurserymen who had heard of the St Louis codes three years after their ratification.³⁶⁵⁷

Duties on landowners

In England and Wales it is an offence to plant giant hogweed or otherwise cause it to grow in the wild, but not to have it in a garden.⁵⁸²² Since 12 April 2015 an authority can require its control through a species control agreement or order and, according to the RHS, do the work itself and recover the costs.⁵⁹²² In Latvia owners must register infestations, carry out a control plan and monitor the land for five years, and a flowering plant with a developed umbel counts as proof that control was not carried out.¹² In Russia holders of land have had to destroy “dangerous invasive” plants since 1 March 2026,¹⁴ and in Ontario owners whose noxious weeds harm farmland are responsible for their control and its cost.¹³ Swedish owners are in principle obliged to act, but no management programme for giant hogweed had been adopted, because Naturvårdsverket, the environmental protection agency, found the law unclear about what the measures were and who was responsible.⁴⁰

Polish law in 2014 prohibited knowingly keeping or cultivating the Caucasian hogweeds, but put no duty on owners to destroy them: an owner who showed that they had not known of the plants escaped liability, so that simple neglect could not be punished.⁸not knowing was a defence Since 2 August 2017 giant hogweed may not be placed on the market, grown or kept in the European Union.⁶⁰⁶¹

Control on private land needs consent. New York’s crews must obtain the owner’s permission to enter and treat private land; in 2023, 126 active sites went without monitoring or control, 84 per cent of them for want of contact with the owner or of permission, and permission was refused at 33.⁶²the plant needs no permission In Tromsø, Persian hogweed was more likely to be controlled or absent on inhabited parcels, especially those lived in by their owners, and less likely on parcels managed jointly by several households.⁶³

Public programmes and private money

New York’s Giant Hogweed Program, which has run since 2008, counted 1,369 sites eradicated by 2024, and is funded in part by the state’s Environmental Protection Fund and the US Forest Service.⁶⁴ In the Catskills residents are asked not to tackle the plant themselves: the regional partnership or state staff will control it “for free”.⁶⁵ In Poland, in 2014, municipal budgets for control were described as matched to limited means rather than to need.⁸ Some of the money is private. The British booklets are paid for by water companies,² and the clearance of the Kurgalsky reserve by Nord Stream 2 AG.⁴¹

See also

• Control of giant hogweed § After control
• Control of giant hogweed § Grazing
• Enemy release hypothesis § The search for biological control
• Sosnowsky’s hogweed
• Seed bank
• Planned eradication