H. sosnowskyi is a monocarpic biennial or perennial plant that grows up to 3 m tall. Leaves are divided into segments and can reach up to 3 m in length. White or pinkish flowers are organized in umbels and flower...
H. sosnowskyi is a monocarpic biennial or perennial plant that grows up to 3 m tall. Leaves are divided into segments and can reach up to 3 m in length. White or pinkish flowers are organized in umbels and flowering occurs in the second year or later (Nielsenet al., 2005). An average plant can produce around 9000 fruits (Tkatschenko, 1989). The mother plant dies after producing seeds, however the plant is reported to live up to 6 years before flowering (Satsiperova, 1984). The whole plant contains photosensitizing furanocoumarins that can burn and/or blister human skin (after contact with plant sap and subsequent UV irradiation).
The plant’s large size, high fecundity, early germination and vigorous growth makes H. sosnowskyi a very successful invader that can out-compete local flora, cause river bank erosion and is toxic to humans.
H. sosnowskyi had been cultivated for biomass and silage production in the former USSR in the second half of the twentieth century. From the 1990s there were reports of the plant escaping from cultivation and methods of treating H. sosnowskyi as a weed appeared (Polyanski, 1990). Although planting schemes were eventually abandoned, the plant escaped and naturalized and is now widely distributed in the Baltic area and large areas of European Russia (Nielsenet al., 2005). The plant is listed as invasive in several European lists and databases, e.g. EPPO, NOBANIS, DAISIE.
Heracleum sosnowskyi Manden. was described in 1944 (Mandenova, 1944) from Meschetia district in Georgia. It was placed into section Pubescentia Manden. that comprises 13 Heracleum species from the Caucasus, Transcaucasia, Crimea, Persia and Pamiro-Alai (Mandenova, 1950; Jahodová et al., 2007a). According to Mandenova, H. sosnowskyi is similar to H. pubescens (Hoffm.) M. Bieb., which she regarded as endemic to Crimea. Later authors consider H. sosnowskyi a synonym of H. wilhelmsii Fisch. et Avé-Lall. (Grossheim, 1967) or H. wilhelmsii auct. non Fisch. et Avé-Lall. a synonym of H. sosnowskyi (Menitsky, 1991). Recently, A. V. Yena and E. S. Kraynyuk speculate that H. sosnowskyi is synonymous with H. pubescens (Greuter and Raus, 2007) but more research needs to be done to confirm this.
The section Pubescentia comprises other Heracleum species that have been reported as invasive in Europe and/or North America: H. mantegazzianum Sommier & Levier and H. persicum Desf. ex Fisch. Recent molecular-genetic studies showed close genetic relationship between all three invasive species, particularly between H. sosnowskyi and H. mantegazzianum (Jahodováet al., 2007b; Logacheva et al., 2008). However, not all species from the section Pubescentia were analysed in either study.
Because H. sosnowskyi and H. mantegazzianum are not only genetically but also morphologically very similar, several botanists consider H. sosnowskyi only a subtaxon of H. mantegazzianum or H. pubescens. Therefore, H. sosnowskyi does not appear in the lists of weedy flora in many (Western-) European countries (Kabuce, 2006). For practical reasons in countries where both species invade, authorities often collect combined data on distribution and monitoring of these two species (personnel are not trained to distinguish the two species, which requires specialist expertise).
H. sosnowskyi has been bred and hybridized with other Heracleum species. Popular varieties were ‘Uspekh’ and ‘Severzhanin’ (Boodiaket al., 1981; Satsiperova, 1984; EPPO, 2008).
As well as Sosnowskyi’s hogweed, other common names have been used for H. sosnowskyi in English. These include giant hogweed, giant cow parsnip (normally used for H. mantegazzianum), cow parsnip (the common name of North American H. lanatum [H.sphondylium subsp. montanum]) or cow parsley (the common name of Anthriscus sylvestris).
H. sosnowskyi is monocarpic biennial or perennial plant. Height in the invaded range is often reported as 100-300 cm, although in the original description from the native range it is ‘only’ 100-150 cm (Mandenova, 1950). The stem is ridged and sparsely hairy with purple blotches. Leaves of mature plants are divided to a varying extent, either into three approximately equal parts, which may themselves be similarly divided (ternate), or into more than three leaflets arranged in rows along the central leaf stalk (pinnate). The leaf margins have short rounded teeth. On the upper surface the leaves are hairless and below slightly hairy. The flowers are white, sometimes pinkish. Outer petals are radiate, 9-10 mm long. Flowers are organized in slightly convex compound umbels 30-50 cm across. Umbels have 30-75 rays with only short hairs. Flowering typically lasts from June to August. The fruits are egg-shaped or oval mericarps; 9-16 mm long, 5-9 mm wide, densely hairy when unripe. Ripe fruits have wings with numerous spines situated on small spherical or ovoid swellings. Fruits have very conspicuous oil ducts that do not reach the fruit base (Nielsenet al., 2005).
The type specimen was collected in Georgia, in ‘Meschetia, distr. Adygeni in pratis silvaticis in itinere ad jalas Lelovani, 9/8/1936’, and is deposited in Tbilisi (TBI).
H. sosnowskyi is native to the eastern main Caucasian ridge and south-western and eastern Transcaucasia (Mandenova, 1950). Countries in this region that have H. sosnowskyi among their native flora are Georgia, Russia, Armenia, Azerbaijan and Turkey. Type locality lies in Meschetia district, Adygeni in Georgia.
In Europe the plant is distributed mainly in eastern parts - reflecting the history of planting in the former USSR. Most invaded areas are Estonia, Latvia, Lithuania. There are reports of H. sosnowskyi being invasive in Belarus, Hungary, Poland, Ukraine and European parts of Russia, however, detailed distribution for those areas is not known. In Denmark there is only one known population from Ryvangen Naturpark in Copenhagen (Jahodová et al., 2007a). In Germany H. sosnowskyi was grown in trials as a potential fodder crop in the 1960s and there are one or two populations near Steinhöfel-Heinersdorf (east of Berlin) that have survived until now (Zimmermann, 1966; J. Thiele, Institute of Landscape Ecology, Münster, Germany, personal communication).
The distribution in this summary table is based on all the information available. When several references are cited, they may give conflicting information on the status. Further details may be available for individual references in the Distribution Table Details section which can be selected by going to Generate Report.
After World War II (in 1946-1947) the seeds of H. sosnowskyi were brought to the Polar-Alpine Botanical Garden-Institute (PABGI) in Kirovsk, Northern Russia, where experiments took place to investigate the usefulness of this plant as fodder (silage). In 1953 similar experiments began in Leningrad (St Petersburg) - in cooperation with three institutes (the Komarov Botanical Institute and two agricultural institutes). The aim of the experiments and breeding was to produce highly productive cultivars that would have minimal furanocoumarin content.
Seeds of H. sosnowskyi for the first experiments were obtained from Kabardino-Balkaria (North Caucasus, Russia), however later reports mention seeds from other sources, e.g. Dagestan (East Caucasus, Russia) (Marchenko, 1954; Satsiperova, 1984; Jahodováet al., 2007a). Breeding programmes later spread to many other places in the former USSR (including Belarus, Estonia, Latvia, Lithuania, Ukraine). Reports of introduction to these regions do not detail whether seeds were distributed only via the Leningrad and/or Kirovsk institutes or whether additional seeds were obtained from the native range of H. sosnowskyi.
Cultivation on experimental and other farms began soon after the first trials. By 1978 two promising cultivars were bred: ‘Uspekh’ and ‘Severzhanin’ using mass selection. Other methods such as chemical mutagenesis and inter-specific hybridizations were also used in experiments. Other Heracleum species were tested in either monocultures or to make crosses with H. sosnowskyi or other tested species (Dotsenkoet al., 1980; Boodiaket al., 1981; Satsiperova, 1984; EPPO, 2008). Therefore it is possible that hybrids and/or other species that had been tested or planted may have escaped.
Other reasons for introduction included honey production, ornamental purposes and potential medicinal use.
In 1958 studies aimed to determine remedial properties of H. sosnowskyi began in the garden of Therapeutic Plants of Medical Academy in Wroclaw. At the end of 1950s the plant was also introduced as a potential fodder plant
The invasive nature of H. sosnowskyi has been recognized and the plant is now on several lists of invasive species of Europe (EPPO, DAISIE, NOBANIS). There are also several programmes to control and/or manage the plant and to prevent its further spread. However, it is not clear how seriously this problem is seen in Russia as very few records come from this country (as an exception see Filatova and Vlasov, 2002) and some researchers involved with H. sosnowskyi may not accept that this plant can behave as a weed or pest.
Cultivation of H. sosnowskyi for fodder has been abandoned in most places partly because the anise scented plants affect the flavour of meat and milk from animals which eat them and also partly because of the health risk to humans and cattle. However, in parts of northern Russia, agricultural production continues to this day and this can be seen as a potential risk if introduced to areas where it has not been previously cultivated. Moreover, there are reports of new research on the plant for use in the cellulose industry and/or for bio-utilization of slime in Russia, which would presumably require cultivation at farms on a large scale, again posing a risk of introduction and potential escape and invasion into new areas (Nielsen et al., 2005; Musikhin and Sigaev, 2006; N.Timofeev, personal communication).
Due to increasing knowledge of H. sosnowskyi's invasive behaviour and of the toxicity of the plant among authorities and the general public, the risk of deliberate introduction to new areas as an ornamental is low (and is decreasing).
Accidental introductions by humans are still possible (e.g. when seeds become stuck to car tyres and are transported along roads, or when people collect dry flower heads for decorative purposes).
H. sosnowskyi is mostly found in artificial habitats (roadsides, disturbed areas, agricultural fields, abandoned farmyards and gardens) and semi-natural habitats (bushes, grasslands, parks, pastures, abandoned orchards). H. sosnowskyi rapidly invades not only open areas but also spaces along water basins, roads and forests. Its naturalization is favoured in abandoned land, where earlier land management activities took place (Laivinš and Gavrilova, 2003; Kabuce, 2006; Obolevica, 2009).
In natural habitats H. sosnowskyi develops large stands, for example in meadows, river valleys and fringes of forests as well as on flood plains of rivers and lakes. Overall it is typical that the hogweed stands are closed neophyte societies where it is a vigorous dominating species. Hogweed overwhelms native species in occupied areas and therefore hogweed societies are poor in biodiversity (Laivinš and Gavrilova, 2003; Kabuce, 2006). For example, the habitat type colonized by H. sosnowskyi is known for 156 localities in Latvia: the number of localities is higher in cultivated areas, whereas in wooded areas H. sosnowskyi can be found less frequently. H. sosnowskyi has spread mostly along roads – 70 sites (45% of the total number); in grasslands (meadows, lawns, pastures) – 33 sites (21%); and ditches (7%). There are sites also in other habitats – in embankments of rivers and lakes (7), gardens (6), along railtracks (5) and city streets (4), fallow (4), fields (4), weedy areas (4), yards (3), and there are single sites registered in quarry, brushwood, forest fringe, square and park (Laivinš and Gavrilova, 2003).
H. sosnowskyi is a diploid species with 2n=22 (Bell and Constance, 1966). Genetically it is very close to H. mantegazzianum and also to H. persicum. Invasive populations in countries of Europe where the plant was cultivated are genetically close and indicate introduction from the same source (perhaps via one of the breeding stations in Russia). Separate introduction was detected in a population in a park in Copenhagen, Denmark that clustered away from other European populations, together with native populations from Armenia. It is important to note that in this study, limited samples from native ranges were available for analysis (only from Armenia, thus not including samples from the presumed origin of crop plants - in Kabaldino-Balkaria or Dagestan in Russia) (Jahodováet al., 2007b).
Breeding H. sosnowskyi for use as a fodder plant involved methods such as mass selection, chemical mutagenesis and interspecific hybridizations. It is not known whether products from all the methods were later used in large-scale farming. Other Heracleum species involved in hybridizations include: H. lehmanianum, H. trachyloma, H. ponticum, H. dulce and H. asperum (Dotsenko et al., 1980; Boodiak et al., 1981; Demidov and Satsyperova, 1989). Recent hybridization with other invasive H. mantegazzianum plants and/or with native H. sibiricum [H. sphondylium subsp. sibiricum] and H. sphondylium may be possible, but has not been confirmed.
H. sosnowskyi is a monocarpic short perennial (2-6 years) however there are some reports of it being a polycarpic perennial (flowering more than once before dying) (Kabuce, 2006).
Seeds germinate in early spring after morphophysiological dormancy is broken by the cold and wet conditions of autumn and winter. There are discrepancies on the length of the stratification period as Nikolaeva et al. (1985) reports 4-6 months, but Moravcová et al. (2007) report 2 months or less. Autumn germination under favourable conditions would be also possible though has not been observed. Experiments in the Czech Republic have shown that the seed bank is very quickly depleted by rapid germination in spring and seeds do not survive for more than one season. However, experiments in regions where H. sosnowskyi is native or invasive are needed to verify this. Germination of H. sosnowskyi seeds under laboratory conditions is very high (71-94% under different temperature regimes) (Moravcováet al., 2007).
Seedlings emerge early in spring and at this stage they are cold resistant and light demanding. During the first year(s) the plant develops one or several leaves to form a leaf rosette. In the second or third year (or later, depending on the accumulated resources, competition, etc.) the plant forms a flowering stem. The transition to regenerative phase is characterized by increased pinnating of leaves (Satsiperova, 1977). Flowering occurs from June to August and flowers are insect pollinated, and visited by a wide range of insects, including a number of Hymenoptera, Diptera and Coleoptera. The plant is reported to live up to 11 years when planted for biomass and silage production (Satsiperova, 1984).
An average plant of H. sosnowskyi is reported to have produced 8836 fruits in the Leningrad area, Russia (Tkatschenko, 1989). The majority of seeds (98.2%) are distributed in the upper soil layer of 0-5 cm, with little in the deeper layers of 6-10 cm and 11-15 cm (Moravcová et al., 2007). Seeds are dispersed locally near the parent plants, or over longer distances by watercourses.
Physiology and Phenology
H. sosnowskyi reproduces exclusively by seeds, the majority of which germinate in large numbers early the following spring. Young seedlings are fairly cold resistant and can survive temperatures of -4 to -7°C, but from their second year, plants can survive temperatures of -25°C and down to as low as -45°C under snow cover (EPPO, 2008; Obolevica, 2009).
H. sosnowskyi communities have been studied in Latvia and they differentiate into four groups of derivate plant communities: Ds. H. sosnowskyi [Galio-Urticetea], Ds. H. sosnowskyi [Artemisietea], Ds. H. sosnowskyi [Stelarietea mediae], Ds. H. sosnowskyi [Molinio-Arrhenatheretea]. The dominant species are Dactylis glomerata, Urtica dioica, Aegopodium podagraria, Taraxacum officinale, Artemisia vulgaris and Anthriscus sylvestris (Laivinš and Gavrilova, 2003).
In a study in Latvia, Laivinš and Gavrilova (2003) report that derivate communities of H. sosnowskyi develop in fresh and slightly moist neutral soils rich in nutrients. Average values of pHKCl vary from 6.3 to 7.0, base saturation from 91 to 98%, and C/N ratio from 6 to 9. H. sosnowskyi rarely occurs in locations where pH is lower than 6.0. In H. sosnowskyi communities, topsoil is rich with organic matter and nitrogen. Mineralization of organic matter in agricultural lands in H. sosnowskyi communities is very intensive and the C/N ratio is less than 10; in ruderal habitats this process is slower.
Studies on natural enemies of H. sosnowskyi were performed mainly in the period when this plant was cultivated and thus the natural enemies were treated as pests. Few studies later on mention the potential for biological control and no reports in the literature were found to indicate intentional use of any biocontrol agents.
One study was conducted in the native range (in Turkey) and found a weevil, Lixus nordmanni, that feeds and grows on some species of Apiaceae (H. trachyloma, H. sosnowskyi, H. pastinacifolium, H. platytaenium, Angelicasylvestris and Falcaria sp.). Lixus nordmanni has one generation per year in eastern Anatolia and adults and larvae feed on leaves and stems of the host plant (Gültekin, 2006).
In the introduced range, herbivorous insects and pathogens were recorded as pests and enemies of H. sosnowskyi. Reported insects include:
H. sosnowskyi is naturally dispersed by water and wind. Due to the lack of studies of seed dispersal of H. sosnowskyi and similarities to the well studied H. mantegazzianum, the following information is based on seed dispersal of H. mantegazzianum:
The majority of seeds are released in the immediate vicinity of the parent plant. In individuals of about 2 m tall, 60-90% of seeds fall on the ground within a radius of 4 m of the parent plant. Dispersal by wind is usually only at short-distance (2-10 m), although it can become important at longer distances in winter when seeds are blown over frozen or snow-covered soil surfaces (Nielsenet al., 2005; Kabuce, 2006). Natural longer-distance dispersal is usually achieved by water, as the seeds are able to float for at least 3 days (Pyšek, 1994).
Vector Transmission (Biotic)
Seeds can be transported attached to animal fur (reported for sheep and cattle by Nielsenet al., 2005). No dispersal by birds has been observed (EPPO, 2008).
Seeds can be transported unintentionally by movement of soil during building and excavation, by transport along roads or railways by attachment to vehicles or by air currents and by movement by agricultural and forest tractors which carry seeds on radiators and roofs. Collection of ornamental seed-heads followed by disposal on rubbish heaps can also contribute to accidental introduction to new areas (EPPO, 2008).
Intentional introduction was the main source of the plant in the past. It was cultivated as a fodder crop and for this reason also grown in various trials and experiments. Other intentional introductions included planting by beekeepers to increase honey production, planting in gardens as an ornamental plant and growing as a vegetable used in salads and/or soups. Studies on essential oils and potential medicinal use of H. sosnowskyi may have also added to the spread assisted by humans.
medicinal costs associated with treating people with dermatitis induced by contact with H. sosnowskyi;
management or eradication costs.
The problem is especially severe in Baltic countries, where H. sosnowskyi invasion is out of control (Kabuce, 2006). For example in Latvia the area infested by this species was estimated as 12,182 ha in 2001 with the prognosis that the area will increase to over 18,000 ha in 2007 (Obolevica, 2009).
Detailed quantitative calculation for impact of H. sosnowskyi is not known, however a German study that assessed the economic impact of the closely related H. mantegazzianum shows the cost is more than 12 million euros annually in the country. This sum is distributed among the health system (1,050,000 euros), nature reserves (1,170,000 euros), road management (2,340,000 euros), municipal management (2,100,000 euros) and district management (5,600,000 euros) (Reinhardt et al., 2003).
In the past there was a positive economic impact as the plant was used to produce fodder (silage) for cattle or to increase honey production. But agricultural production has been abandoned (except in parts of northern Russia).
H. sosnowskyi has spread actively in Baltic countries in artificial and semi-natural habitats over the last 20 years. The plants are forming communities that are nitrophilous and the eutrophication of the environment further stimulates their dissemination (Laivinš and Gavrilova, 2003).
H. mantegazzianum has been reported to cause riverbank erosion through the suppression or exclusion of native species, which play an important role in riverbank stabilization. A similar effect is also expected for H. sosnowskyi. When Heracleum plants in dense stands die off in winter, they leave bare soil that can be eroded by rainfall or winter floods. Deposition of eroded silt can alter substrate characteristics in rivers, and, for example, render gravel substrates unsuitable for salmonid spawning (Thiele and Otte, 2007).
Impact on Biodiversity
The enormous height and leaf area of H. sosnowskyi together with its ability to reach high densities in abandoned grasslands and ruderal habitats leads to a strong decline in the species richness of these habitats. In such areas, up to 80% of the incoming light is absorbed by H. sosnowskyi thus suppressing native plant species that may become endangered (Nielsen et al., 2005).
Hybridization poses a possible threat in two respects. One is possible hybridization of invasive H. sosnowskyi with native species (H. sphondylium subsp. sibiricum and H. sphondylium). Although not yet reported, similar hybridization is known to occasionally occur between a close relative H. mantegazzianum and native H. sphondylium in the UK and Germany (McClintock, 1973; Grace and Nelson, 1981; Stewart and Grace, 1984; Ochsmann, 1996). Second, as we know from breeding programmes in Russia, hybridization in the genus Heracleum is possible and hybrids are readily obtained in these programmes (Demidov and Satsiperova, 1989). Therefore, hybridization involving several invasive Heracleum species (e.g. H. mantegazzianum and H. sosnowskyi) cannot be ruled out. This may present a serious threat should such hybrids with invasive traits superior to their parents be formed (Ellstrand and Schierenbeck, 2000). Possible hybrid individuals were detected in the study of Jahodová et al. (2007b) from Latvia.
H. sosnowskyi (as well as other invasive Heracleum species) contain photosensitizing furanocoumarins that cause a phytotoxic reaction (burning, blistering) in human skin after contact with the plant and subsequent UV irradiation. The phototoxic reaction can be activated 15 min after contact, with a sensibility peak between 30 min and 2 hours. After about 24 hours, flushing or reddening of the skin (erythema) and excessive accumulation of fluid in the skin (oedema) appear, followed by an inflammatory reaction after three days. Approximately one week later a hyper-pigmentation (usually darkening the skin) occurs which can last for months. The affected skin may remain sensitive to ultraviolet for years. In addition, several furanocoumarins have been reported to cause cancer (carcinogenic) and to cause malformation in the growing embryo (teratogenic) (Nielsenet al., 2005).
Dense stands of H. sosnowskyi can obstruct access to river banks, lakes, parks and/or forests, thus affecting the recreational value of such land (influencing anglers, water sports enthusiasts, swimmers, bird watchers, hikers etc.). It may also present a road safety hazard by reducing visibility along roads (based on reports for H. mantegazzianum, e.g. Thiele and Otte, 2007).
In the past H. sosnowskyi was cultivated for fodder and to increase honey production, but these practices have been abandoned in most countries. However, in parts of northern Russia agricultural production continues to this day, but the economic value of trials or farm-scale production is not known.
Detection should initially be focused on mapping the distribution of all populations in a country or area of interest. Owing to their large size, stands of H. sosnowskyi are very conspicuous and it is therefore relatively easy to determine the plant’s distribution. Involving the public in locating stands through a public awareness raising campaign is particularly suitable. In cases where more tall invasive Heracleum species occur in a territory of interest, distribution can be mapped together (as distinguishing between the species requires specialist knowledge). Another useful method, which could be used to determine the distribution of these conspicuous aliens, is aerial photography during the flowering and early fruiting period (from the second half of June to July; Nielsenet al., 2005).
H. sosnowskyi is closely related to H. mantegazzianum and can be easily confused with it. According to Mandenova (1950) there are two characteristic and constant features of H. sosnowskyi: the form of the vittae of fruits that are brown and swollen (but quite narrow in H. mantegazzianum); and rays of the umbel have only short hairs (but both long and short flexuous hairs in H. mantegazzianum). Other features distinguishing H. sosnowskyi include leaf margins that have short rounded teeth (but a saw-edged margin in H. mantegazzianum), slightly convex umbel (as opposed to a flat umbel in H. mantegazzianum), fruit wings with numerous spines situated on small spherical or ovoid swellings (while H. mantegazzianum has fruit wings hairless or with solitary spine-like hairs). Although H. sosnowskyi usually has few divided leaves with more rounded segments, the shape of leaves is extremely variable in both species and cannot, on its own, be used in species identification.
H. persicum is another invasive Heracleum species that can be mistaken for H. sosnowskyi. In contrast with the previous two species it is a polycarpic perennial plant (it flowers and produces seeds more than once). Leaves are pinnately divided in five segments, the stem is often purple and the whole plant has an anise odour.
The native Heracleum species in Europe (the most common being H. sphondylium and H. sibiricum [H. sphondylium subsp. sibiricum]) and in the USA (H. sphondylium subsp. montanum) should not be easily confused with H. sosnowskyi as they are much smaller, rarely over 2 m high, with grey-green, pubescent and less acutely toothed leaves.
There have been a few reports that some other species from the carrot family (Apiaceae) have been mistaken for H. mantegazzianum. Because of the close relationship between these two Heracleum species, occasional confusion between the following species and H. sosnowskyi cannot be ruled out:
Wild parsnip (Pastinaca sativa) is distinguished by its yellow flowers and hairless leaves that are simply divided into more than three leaflets with a v-shaped base.
Wild angelica (Angelica sylvestris) is almost hairless and has characteristic purple bands at the base of the leaf and leaflets. Oil ducts of fruits are small and reach the base.
Garden angelica (A. archangelica) is a common garden plant that is cultivated for its aromatic stems and oils distilled from the seeds and root. The plant grows to 100-230 cm tall, leaves and stem are hairless. Umbels are strongly convex with greenish flowers and the egg-shaped fruits are without conspicuous oil ducts.
H. sosnowskyi has been recognized as an invasive plant by several European organizations (EPPO - http://www.eppo.org, DAISIE - http://www.europe-aliens.org, NOBANIS - http://www.nobanis.org), who also recommend management plans. A best practice manual has also been published by members of the EU-funded Giant Alien project (http://www.giant-alien.dk).
Specific management and or eradication programmes are also run on regional and local scales. For example the Ministry of Agriculture of the Republic of Latvia has provided financial support to elaborate the Provisional Recommendations to localize the Heracleum species in Latvia. In the National Programme of Biodiversity, it was acknowledged that H. sosnowskyi is a very expansive species that can invade and establish not only in weedy places and on road margins, but also in natural plant communities (Obolevica, 2009). In 2005, the Ministry of the Environment of the Republic of Latvia prepared and accepted the National Hogweed Limitation Programme 2006-2012, which aims to reduce H. sosnowskyi as an invasive weed in Latvia. The programme proposes establishment of a coordinated hogweed limitation system – including the establishment of a responsibility and punishment system, administration of financial resources, as well as development of regulation regarding beneficiaries involved in management efforts (Kabuce, 2006).
Preventive measures should include the prohibition of seed importation and the regulation of habitat quality in areas prone to invasion. To reduce seed input into unoccupied areas in the first place, sowing and planting of H. sosnowskyi in gardens, parks and the open landscapes should be prohibited. Additionally, unintentional transport of seeds, e.g. through transfer of soil material, should be avoided. Extant stands along transport habitats (water courses, highways) should be managed to prevent dispersal of seeds.
In areas prone to invasion, management should be continued for as long as possible in a regular fashion, e.g. annually, and with sufficient intensity. Mowing and grazing are suitable for the management of grasslands and grassland-like fringe habitats. Abandoned fields should be kept under observation as they can be easily invaded, especially areas with patches of exposed soil.
High risk points of entry should be identified and visited. These include parts of a landscape that are:
- within wind dispersal distance of existing stands, but not protected by barriers (e.g. dense scrubland and forest);
- within the flood zone of water courses, where tall invasive species of hogweed occur in the upstream flood area;
- road or railway borders within 2 km of another invasive hogweed patch along the border;
- adjacent to gardens in which invasive hogweeds are cultivated.
Regular checks of these areas could be combined with monitoring for other alien problem plants (Nielsenet al., 2005).
Early warning systems and rapid response
If prevention methods fail, early detection of new populations is essential to facilitate a rapid eradication response. Mainly because eradication efforts are cheaper and more likely to be successful while new infestations are still small. Early detection, however, will only be useful if it is backed up with a contingency plan for eradication. This plan needs to identify responsible organizations or groups committed to take rapid action, and to ensure that they have appropriate financial, human and material resources. Actions will only be successful, if they are embedded in management plans integrated across different regional and local authorities (Nielsenet al., 2005).
The general public can be informed of the problems caused by H. sosnowskyi and asked to help identify populations and/or single plants of invasive hogweeds and to help with containment and eradication programmes. This can be done using internet sites for a particular area, local papers, radio and television programmes and posters, brochures and leaflets. A focused programme of awareness raising needs to be developed: targeted at key groups, e.g. road and river managers and companies deliberately or inadvertently transporting soil. Groups involved in outdoor activities, such as fishermen, farmers, hunters, environmental groups, hiking and cycling clubs, can be directly targeted. The public needs to be aware (or easily able to find out) where their observations should be reported (Nielsenet al., 2005).
Eradication of H. sosnowskyi can be achieved using several control techniques (such as manual and mechanical methods, grazing or application of herbicides) or a combination of these. Subsequently, the infestation area, including adjacent areas that might have received seeds (e.g. downstream from a riverside infestation), should be monitored and any regrowth or new infestations eradicated. Since the seeds can survive for several years in the soil, follow-up monitoring of the eradication site for at least five years must be undertaken both within and outside the area to be protected. Plants in the first year are more difficult to find, so this should be undertaken by staff familiar with the plant in its vegetative stages (Nielsenet al., 2005).
Cultural control and sanitary measures
Intensive grazing has proved to be very efficient for the control of large stands of invasive Heracleum species, especially those inaccessible by machinery. Experience with livestock grazing has been gained mainly from the use of sheep grazing on H. mantegazzianum, but it is believed this control method is also applicable to other tall invasive Heracleum species, including H. sosnowskyi. Heracleum spp. are also very palatable to cattle, and goats and pigs have been reported to graze on them too (Buttenschøn and Nielsen, 2007)
Sheep were found to greatly reduce H. mantegazzianum after 2 years and completely eliminated it after 5 years when no viable seeds remained in the soil (Andersen and Calov, 1996). Generally, grazing promotes a dense sward of grazing-tolerant species and limits the amount of suitable ground in which seeds of Heracleum spp. can germinate and become established.
Sheep and cattle prefer young and fresh plants, and the most efficient control is obtained by beginning the grazing early in the season when the plants are small. It is recommended to use a dense regime of animals in spring (20-30 sheep/ha), and reduce grazing pressure at the end of June (5-10 sheep/ha) when the plant is weakened and most of the plant biomass has been removed. Grazing is a cheap method when large fenced areas can be established but should also be considered for smaller stands if neighbouring areas are grazed and livestock can be relatively easily transferred for shorter periods. If possible, the fenced area should not only include the colony of tall invasive species of hogweed but also the surrounding area where seed dispersal may have taken place. (Nielsenet al., 2005).
Furanocoumarins present in sap of Heracleum spp. that causes blistering in humans can sometimes also affect grazing animals. Therefore choosing livestock with pigmentation of the bare skin, e.g. blackfaced sheep, can reduce inflammation and/or blistering around the mouth, nostril, eyes, ears etc. Affected animals must be removed from the field temporarily. Clinical studies showed a reduced fecundity after oral application of furanocoumarins. However, this phenomenon has not been reported so far for grazing animals (Nielsenet al., 2005; Buttenschøn and Nielsen, 2007).
Manual and mechanical methods to control H. sosnowskyi include techniques such as root cutting, cutting the plant, mowing, umbel removal and ploughing. Except for root cutting, mechanical control does not cause the immediate death of the plants. Death occurs after 2-3 treatments per year during several growing seasons through depletion of nutrient reserves.
Root cutting (or digging) is effective but labour intensive and is recommended for single plants or small stands <200 plants). It is usually performed with an ordinary spade with a sharpened blade. The cutting should take place in early spring and be repeated in mid summer. It is recommended that the root is cut at least 10 cm below soil level (or up to 25 cm in cases when additional soil layers may cover the plants).The cut parts of the plants are pulled out of the soil and either destroyed or left to dry out. The cost of one treatment in the first year with an estimated rate of control of 100 plants/hour was 36,894 euros (Nielsenet al., 2005).
Mechanical mowing techniques, for example a flail mower, are useful for large infested areas. The plants will rapidly recover with basal regrowth and mowing must be repeated 2-3 times during the growing season to hinder the re-sprouting plants from storing nutrients in the root and then flowering and setting seed. If the population is small or situated in a location unsuitable for mechanical mowing, e.g. along rivers or on slopes, the plants can be cut manually by using a scythe or a trimmer. One strategy of cutting is to cut only the flowering plants at mid-flowering stage. Production of new seeds will be prevented and vegetative plants will out-shade each other. When repeated carefully this strategy should with the least efforts eradicate the population in a few years. The cost of three treatments in the first year with an estimated rate of control of 500 plants/hour was 22,176 euros (Nielsenet al., 2005; Pyšek et al., 2007).
Removal of umbels can be as effective as cutting the whole plant, but this control method often fails to prevent seed production due to the high regeneration potential. Plants subjected to removal of flower heads can readily regenerate and produce new flowers and viable seeds of normal size that germinate well. This method should only be considered as an improvised solution for control of stands where no other attempts of control have taken place earlier in the season. Timing of removal is crucial because if the treatment is applied too early in the season (before full inflorescence), regeneration is very vigorous and an even larger number of seeds is produced. If treatment is too late (at the beginning of seed-setting), there is a risk that seeds will ripen even on cut umbels that are left lying on the ground. The cut umbels must be collected and destroyed. The removal of umbels is most effective if done when terminal umbels just start to flower. Even then, there is some regeneration and treated stands must be checked at the time of seed ripening to prevent release of seeds produced by regeneration (Nielsenet al., 2005; Pyšeket al., 2007).
Ploughing can control an infestation of H. sosnowskyi on agricultural land. Deep ploughing of the soil (up to 24 cm) will significantly reduce the germination of hogweed seeds due to the upper soil (where the majority of the seeds are concentrated) being buried. The best results are obtained if the established vegetation of invasive hogweed plants is controlled mechanically or chemically prior to the ploughing (Nielsenet al., 2005).
No agent (insect nor pathogen) able to cause considerable damage and to be specific enough to be considered safe, has been found for any invasive Heracleum species so far.
H. sosnowskyi (as well as other invasive Heracleum species) is susceptible to systemic herbicides such as glyphosate, triclopyr and imazapyr. Glyphosate is the most widely used compound, but owing to risks of toxicity to fish and algae, an unsprayed buffer zone of 2 m should be left adjacent to any river or other water body (Marcher, 2001). Triclopyr has no effect on germinating grasses and is useful in controlling a range of broad-leaved species. Imazapyr has a residual effect in the soil that prevents further germination but may also impact on non-target species (EPPO, 2008).
EU policies and national rules and guidelines should be consulted before any herbicidal application is undertaken. Spraying should be done in dry and calm weather. If herbicides are to be applied it is strongly recommended that the plants are treated early in spring when they have reached a height of 20-50 cm and access to the centre of the colony is still possible. A follow-up spraying may need to be performed before the end of May to target new seedlings. This second application may be replaced by mowing or cutting. The estimated cost of two treatments in the first year by hand held equipment with an average of 300 m² sprayed/h was 20,196 euros (but note that the costs of chemicals and equipment were not included in the calculation; Nielsenet al., 2005).
Integrated pest management strategy is preferred and should focus on optimal management with respect to efficacy, ecology and economy. Such a programme should contain clearly defined objectives (e.g. eradication or containment), identification of all stands in the area and habitats prone to invasion and availability of resources. The selection of control method(s) depends on many factors and often a combination of two or more methods can be more efficient. Regardless of the control method, management usually requires repeated and correct application in order to obtain satisfactory control. Accordingly, to make the most of the control efforts, the treatment of plants should be started early in the growing season and continue for several years until the soil seed bank is depleted and the root system has died. Due to the likely spread of seed down a river, it is important to ensure that upstream infestations are dealt with before attempting eradication further down (Caffrey and Madsen, 2001; Nielsenet al., 2005, 2007; EPPO, 2008).
Monitoring and Surveillance
Monitoring is an important part of the integrated management strategy throughout the management process. It starts with the detection and mapping of invaded areas which helps to define management and control plans. During the control and eradication programmes monitoring is a useful tool to evaluate success of these programmes and direct efforts to problematic areas. Following any eradication programme it is essential to monitor the areas for potential regrowth from seedbank and/or possible re-introduction from neighbouring regions for rapid response.
Due to the size and conspicuous nature of H. sosnowskyi there are various options for monitoring, including aerial photography and involving the general public or interest groups.
Revegetation of areas invaded by H. sosnowskyi should be an essential part of any eradication or control programme. As H. sosnowskyi was grown as an agricultural crop, the majority of land invaded is therefore abandoned arable land, meadows or pastures. The integrated control using mowing/cutting, chemical control, soil cultivation and sowing of grass mixtures is recommended (Nielsen et al., 2005; Ravn et al., 2007).
Grass mixtures should be sown at high seed densities (4,000 emerging seedlings/ m2). Native grass species and cultivars must be chosen that have proven to be competitive, produce dense swards, are suitable for growing in mixtures, and make good growth after repeated cutting. Examples of grass mixtures that have proven suitable are: Dactylis glomerata, Festuca rubra (50:50), and Lolium perenne, Festuca rubra, Poa pratensis (12:35:53). A selective herbicide suitable for broadleaved weeds in the developing grass sward (including newly emerged seedlings of invasive H. sosnowskyi species) may be used as a single application during the vegetative period (Nielsenet al., 2005).
In natural habitats, e.g. along riversides, it is not advisable to apply herbicide treatments prior to the sowing of grass mixtures. The creation of a strong competitive plant community for the depression of H. sosnowskyi and stability of the soils against erosion is achieved by additional cutting treatments and increasing sowing rates of grass mixtures.
In places with high densities of H. sosnowskyi, above-ground cutting of all plants in the spring is recommended after the over-wintering plants have resprouted. Sowing grass mixtures at increased sowing rate and with the ability to grow well after frequent cutting is recommended for re-establishment of native grass sward. Seeds may be sown by hand. The best seeds for this purpose are those types of grass that are very abundant in the locality, resistant to flooding, well-adapted to the habitat and able to compete with H. sosnowskyi. Examples of grass mixtures that have proven suitable are Dactylis glomerata, F. rubra (50:50), and F. arundinacea, Festuca rubra (35:65). Frequent cutting of the re-established grass sward is recommended, when the height of H. sosnowskyi seedlings reaches 20-30 cm (Nielsenet al., 2005).
Sowing grass mixtures in sandy soils of a floodplain and along riverbanks should be avoided, because the seeds of invasive H. sosnowskyi species along with other seeds are deposited by flooding. The seeds are retained in the grass sward and will germinate well in spring. In such places only root cutting and cutting of the plants before flowering can be recommended for the control of invasive H. sosnowskyi (Nielsen et al., 2005).
Afforestation is a special case of a cover crop strategy that has been suggested for Heracleum species, however H. sosnowskyi is more shade tolerant (and therefore less suitable for this treatment) than for example H. mantegazzianum. There is also a variation among trees in shading out ability. Fagus sylvatica is more suitable in this respect than Abies and Salix species (Nielsenet al., 2005).