Ecology, growth and management of black locust (Robinia pseudoacacia L.), a non-native species integrated into European forests

Black locust (Robinia pseudoacacia L.), a species native to the eastern North America, was introduced to Europe probably in 1601 and currently extends over 2.3 × 106 ha. It has become naturalized in all sub-Mediterranean and temperate regions rivaling Populus spp. as the second most planted broadleaved tree species worldwide after Eucalyptus spp. This wide-spreading planting is because black locust is an important multipurpose species, producing wood, fodder, and a source of honey as well as bio-oil and biomass. It is also important for carbon sequestration, soil stabilization and re-vegetation of landfills, mining areas and wastelands, in biotherapy and landscaping. In Europe, black locust is drought tolerant so grows in areas with annual precipitation as low as 500–550 mm. It tolerates dry, nutrient poor soils but grows best on deep, nutrient-rich, well-drained soils. It is a fast-growing tree and the height, diameter and volume growth peak before the age of 20. It mostly regenerates vegetatively by root suckers under a simple coppice system, which is considered the most cost-effective management system. It also regenerates, but less frequently, by stool sprouts. Its early silviculture in production forests includes release cutting to promote root suckers rather than stool shoots, and cleaning-respacing to remove low-quality stems, reduce the number of shoots per stool, and adjust spacing between root suckers. In addition, early, moderate and frequent thinning as well as limited pruning are carried out focusing on crop trees. The species is regarded as invasive in several European countries and its range here is expected to expand under predicted climate changes.

wood, fodder, and a source of honey as well as bio-oil and biomass. It is also important for carbon sequestration, soil stabilization and re-vegetation of landfills, mining areas and wastelands, in biotherapy and landscaping. In Europe, black locust is drought tolerant so grows in areas with annual precipitation as low as 500-550 mm. It tolerates dry, nutrient poor soils but grows best on deep, nutrient-rich, well-drained soils. It is a fast-growing tree and the height, diameter and volume growth peak before the age of 20. It mostly regenerates vegetatively by root suckers under a simple coppice system, which is considered the most cost-effective management system. It also regenerates, but less frequently, by stool sprouts. Its early silviculture in production forests includes release cutting to promote root suckers rather than stool shoots, and cleaning-respacing to remove low-quality stems, reduce the number of shoots per stool, and adjust spacing between root suckers. In addition, early, moderate and frequent thinning as well as limited pruning are carried out focusing on crop trees. The species is regarded as Introduction Ecology, growth and management of black locust (Robinia pseudoacacia L.), a non-native species… areas of Germany, this species has produced 3-10 oven-dry t ha −1 a −1 , substantially higher than poplar and willow clones established on the same sites (Grünewald et al. 2009). In Italy, on poplar sites, its production can reach 10, even 12.5 dry matter t ha −1 year −1 using selected clones of Hungarian provenance (Giulietti 2016). In Hungary, where black locust is considered the most suitable species for energy plantations (Rédei and Veperdi 2009), its productivity reached 9.7 dry matter t ha −1 yr −1 (Rédei et al. 2011). Black locust is also important as a raw material in the food industry, as fodder in silvopastoral systems, in biotherapy, and landscaping (Cierjacks et al. 2013;Enescu and Dănescu 2015;Ciuvăț et al. 2015;FOREST EUROPE 2015;Sitzia et al. 2016;Vítková et al. 2016). It is important in honey production (bees can collect up to 1500 kg of honey from one ha of black locust stands in favourable years ;Vakulyuk 1991), and it provides habitat for numerous fungi, arthropods, birds and mammals (Saunier 2007;Ślusarczyk 2012;Ciuvăț et al. 2015).
In Europe, black locust is cultivated on a large scale for timber production; curiously, it is not an important Mid nineteenth century (Kirov 1886) 1890 (Naydenov 2015) Croatia 1856 (Vuković et al. 2013) 1880 (Đodan et al. 2018) Germany Ca. 1650 (Seeling 1997;Krausch 2001) Ca. 1700 (Seeling 1997;Krausch 2001) Hungary 1710-1720 (Rédei pers.comm.) Beginning of nineteenth century (Rédei pers.comm.) Italy 1662 (Botanical garden of Padua) (Pignatti 1982) Beginning of nineteenth century (Zucchini 1800) Romania Ca. 1750 (Drăcea 1919) 1852 (Crăciunescu 1904;Drăcea 1919) Slovakia 1720 (Benčať pers.comm.) 1780 (Benčať pers.comm.) Slovenia 1790-1800 (Strgar 1992) 1810 (Anko 1989) Ukraine End of eighteenth century (Kohno 2007) Beginning of nineteenth century (Vakulyuk and Samoplavskyy 1998) United Kingdom 1640 s (Savill 2013) Early nineteenth century (Savill 2013)  commercial timber species in native range (Huntley 1990), where the best uses are mine timbers, fence posts, poles, railroad crossties, stakes and fuel (DeGomez and Wagner 2001). The wood is dense (specific gravity about 0.73-0.74 g cm −3 at 15% moisture content , Erdélyi 1988;Savill 2013), even though the species is fast-growing; this combination of high wood density with high growth rate is rare among woody plant species (Hanover 1990). Black locust wood is naturally durable so can be used outdoors without any treatment. For example, it is the only broadleaved species grown in France with a natural durability specific to class 4 of the European/ French Norm NF EN 335-1 and can be used in contact with soil or freshwater and exposed permanently to humidity (Borde 2018). The uses of black locust wood in Europe include general construction, boat building, flooring/parquet, terrace floors, outdoor work, veneer, furniture, mine timbers, barrel staves, railway sleepers, vineyard poles and props, fence posts, boxes, crates, handcrafts, wood chips, pellets, particle board, fibreboard (Erdélyi 1988;Torelli 2002;Cierjacks et al. 2013;Savill 2013;Ciuvăț et al. 2015;Sitzia et al. 2016). The species produces highly-valued firewood, very important in many rural areas of Europe such as in Hungary, Serbia, Romania, Bulgaria, and Croatia. Its calorific power (97%) is comparable to that of native oaks (pedunculate oak, Quercus robur L. and sessile oak, Q. petraea (Matt.) Liebl.), considered as 100% but lower than hornbeam (Carpinus betulus L.) (110%) (CIRAD 2012;Bousse and Richard 2017). The importance of this species in parts of Europe is underlined by black locust breeding and improvement programs, started in the 1950s in Germany (Schneck 2010 in Hungary (Keresztesi 1983) and Romania (Bîrlănescu et al. 1966(Bîrlănescu et al. , 1977, 1970's in Bulgaria (Kalmukov 2014), and in the 1980' in Austria (Müller 1990). These programs included initially the selection of individual trees of good stem form and vigor ('plus-trees') and of seed stands, followed by the vegetative propagation of selected 'plus-trees' and establishment of seed orchards. Using selected clones and cultivars, especially in Hungary, wood production has increased significantly, i.e., by 25-30% in Hungary (Rédei et al. 2017) and 48% in Italy (Gras 1991). Breeding and improvement programs are undertaken in Hungary, Bulgaria, France, Greece and Poland, mainly using European provenance stock (van Loo et al. 2017). The array of breeding objectives is very wide, e.g., improvement of wood quality, increase of biomass production for energy purposes, increase of flower nectar production and extension of blossom period, and improvement of drought and frost tolerance (Gras and Mughini 2009;Kalmukov 2014;van Loo et al. 2017).
However, despite its economic and social value, the role of black locust in European forestry is controversial because of its invasiveness. It is ranked amongst the 100 most invasive alien species ("… a species that is established outside of its natural past or present distribution, whose introduction and/or spread threaten biological diversity"-Convention on Biological Diversity 1992, in https ://www.iucn.org/theme /speci es/our-work/invas ive-speci es) on the continent (Vilà et al. 2009). However, it is formally considered to be invasive only in a relatively low number (13 out of 35 reporting) of European countries . For example, German forest scientists consider black locust as "partly invasive" since its invasiveness is mostly restricted to open fields and does not occur in dense forests, with closed canopies . For Slovenian forest experts, black locust is an invasive species (Rudolf and Brus 2006). This emphasizes the ongoing controversy about the species: black locust has many positive socio-economic effects as shown above but can also have negative environmental effects on native vegetation as a light-demanding pioneer species that can substantially, and rather quickly, change the specific habitats which it invades (Vítková et al. 2016. Furthermore, as noted by Vítková et al. (2017), black locust has been widely accepted by the general public in central Europe, where it features in cultural and traditional artefacts.
Recent reviews [e.g., Cierjacks et al. 2013;Vitkova et al. (2017)] have described many aspects of the ecology of black locust, including ones that influence the invasiveness of the species in different parts of Europe. However, there is no comprehensive summary of the silviculture, productivity and management of black locust in Europe that can be used to inform about its future role in different countries. In this paper, along with the essential characteristics of black locust in Europe (e.g., site requirements, root system, regeneration ecology, invasive potential, shade tolerance, potential for natural pruning, vulnerability to pests and diseases), we summarize such information to outline principles for the management of this species, including stand establishment, early interventions such as release cutting and cleaningrespacing, commercial thinning and pruning (both formative and high), all with the aim of producing different wood assortments.

Species characteristics Climate
In its native range, black locust grows in a humid climate (annual total precipitation 1020-1830 mm), with 150-210 frost-free days (Huntley 1990). In Europe, black locust behaves like a thermophilous species, i.e., it grows well where mean annual temperatures are over 8 °C (Roșu 1996;Rédei et al. 2017) and is very drought tolerant (Claessens et al. 2006;Grünewald et al. 2009;Mantovani et al. 2014;CRPF Nord Pas-de-Calais Picardie 2015). In this respect, black locust can adapt to prolonged drought by reducing water loss through both reduced transpiration and leaf size Ecology, growth and management of black locust (Robinia pseudoacacia L.), a non-native species… (Mantovani et al. 2014), as well as due to an extended root system which access water in deeper soil layers (Grünewald et al. 2009). Consequently, the species can grow in regions with 500-550 mm annual precipitation (Roșu 1996;Mantovani et al. 2014), with reports of it tolerating less than 400 mm year −1 (Engel et al. 2014), high summer air temperatures of 30-35 °C, and relative air humidity in July of 20-50%, as found in Hungary (Rédei et al. 2002).
Late and early frosts, as well as low winter temperatures, are damaging for black locust both in Europe (Haralamb 1967), and in the USA (Huntley 1990). Early frosts damage un-hardened twigs and are the source of frequent forking of black locust (Haralamb 1967;Claessens et al. 2006). Storms, wet, sticky and heavy snowfall, and ice can break branches, trunks, forks or shoots at their junction with the stem (Kitin 1960;Vajda 1974;Stănescu 1979;Brus 2011;Meyer-Münzer et al. 2015). Strong winds can cause damage (breakage, uprooting) which, although very rare, can occur on forked black locust individuals (Jovanović 1967;Herman 1971;Vajda 1974;Rițiu et al. 1988).

Soil and topography
In the United States, black locust is found naturally over a wide range of soils but grows best on moist, rich, loamy soils or those of limestone origin (Huntley 1990). It thrives on moist slopes of the eastern mountains, the Appalachians, below 1040 m a.s.l. (Harlow et al. 1986), while in the Great Smoky Mountains National Park, its upper elevational limit is 1620 m a.s.l. (Whittaker 1956in Huntley 1990. In Europe, the species has been introduced over a wide range of soils and tolerates diverse pH from extremely acid (3.2) to alkaline (8.8) (Vítková et al. 2015). However, it grows best when the soil pH is slightly acid to near neutral, 5.5-7.0 (Roșu 1996;Borde 2011;Zajączkowski and Wojda 2012). Black locust tolerates dry and nutrient poor soils, but grows best on deep, nutrient-rich, moist, uncompacted, well-drained silt loams, sandy loams, or sandy soils (Haralamb 1967;Rédei et al. 2011;Meyer-Münzer et al. 2015). Under these conditions, the root-associated nitrogen fixing Rhizobium bacteria, which fixes up to 75 kg of N ha −1 year −1 (Boring and Swank 1984) or 150 kg of N ha −1 year −1 (Berthold et al. 2005), are able to thrive. The groundwater table should be deeper than 150 cm (Keresztesi 1988a, b;Zajączkowski and Wojda 2012), as nitrogen fixation aerobic Rhizobium bacteria is inhibited by high water table or periods of flooding (Vítková et al. 2015). Rendzinas, calcium-carbonated soils in the upper horizons, poorlydrained and highly compacted (clayey) soils, very dry as well as hydromorphic soils with gley or pseudogley, are not suitable for black locust (Stoyanov 1926;Donchev 1968;Rameau et al. 1989;Claessens et al. 2006;Borde 2011;Boer 2013).
In Europe, black locust has been used across a wide range of altitudes, from sea level to 1640 m a.s.l. in the Southern Alps of Italy (Sitzia et al. 2016), but grows best in plains and hilly areas at altitudes below 600-700 m a.s.l. (Vuković et al. 2013;Šnjegota et al. 2017).
However, the root architecture (depth and width) depends on soil conditions: on deep, loose and moist soils as well as dry sandy soils, the roots are well developed, densely branched, with a taproot/vertical roots 1.5-3 m depth (a maximum 7.9 m in Europe, Schütt 1994; a maximum 6-7.5 m in native range, Harlow et al. 1986). On shallow soils or eroded slopes, there is no taproot but fine flat, widespread horizontal roots close to the soil surface with an umbrella-like shape. As acknowledged by Huntley (1990), the species deep rooting ability may explain why can grow in regions much drier than its native range such as in Hungary, Romania, Bulgaria, and China.
Black locust also regenerates on a much smaller scale by stool shoots arising from dormant buds on the stump crown (the callus zone), or on the lower portion of the trunk; sprouting is intensive and persistent (Brus 2011;Boer 2013).
However, root suckers live longer and are healthier, showing less rot at the same age than stool shoots (Negulescu and Săvulescu 1957;Stănescu 1979).
Black locust produces abundant seed crops with 120-150 kg of seeds ha −1 in an average seed year, beginning from around 6 years of age (Rédei et al. 2001;Rédei 2013a, b, c, d). Good seed crops occur every 1-2 years in both Europe (Haralamb 1967;Rédei et al. 2012;Boer 2013;Los et al. 2014;Meyer-Münzer et al. 2015) and the USA (Huntley 1990). The seeds are light (20-23 g for 1000 seeds) and long-lived, mostly gravity-and wind-dispersed in the vicinity of the parent tree. Long-distance spread of seeds by natural means is not very common (Boer 2013;Vítková et al. 2017); however, they have been found up to 67 m away from the parent tree (Morimoto et al. 2010, cited by van Loo et al. 2017. They may persist in a dormant state in the soil for a long time, with a viability longer than 10 years (Haralamb 1967;Boer 2013) or, in some circumstances, up to 40 years (Toole and Brown 1946, cited by Cseresnyés and Csontos 2012), accumulating in a permanent seed bank (Stone 2009;Filipas 2013). This should be taken into consideration when devising strategies to eradicate unwanted black locust (Boer 2013). However, because the heavy, thick seed coat is impermeable to water (both absorption and loss), germination is restricted (McAlister 1971;Vlase 1982;Branquart et al. 2007) so regeneration from seed is rare (Tănăsescu 1961(Tănăsescu , 1967Boer 2013). Germination rate is low under natural conditions (Branquart et al. 2007), ranging from 3.5% to 16.3% (Roberts and Carpenter 1983) to 22% (Singh et al. 1999, both cited by Cierjacks et al. 2013). However, natural regeneration from seed can be facilitated by seed wounding with heavy machinery (Tănăsescu 1961) or by natural thermal shock. The latter occurs particularly on sand dunes where temperatures of the surface layer reach 60-65 °C in summer (Spîrchez et al. 1962).

Invasive potential
As a N-fixing species, black locust creates irreversible changes in physico-chemical and biological soil properties. It influences the species composition of ground flora under its canopy by favouring ruderal and species-poor neutrophilous weed associations, eliminating oligotrophic and acidophilous ones typical of forests, thereby reducing plant diversity (Kleinbauer et al. 2010;Benesperi et al. 2012;Cseresnyés and Csontos 2012;Dimitrova 2012;Vuković et al. 2013). In invaded stands, as soil pH is significantly lower and soil NO 3 − significantly higher, soil microarthropode communities suffer significant loss of abundance and richness while nematode taxon richness is significantly lower (Lazzaro et al. 2018).
Black locust also produces different allelopathic compounds (e.g., robinetin, myricetin and quercetin) which may affect the ecological functioning of natural areas by inhibiting the growth of various weeds and crop species (Nasir et al. 2005;Boer 2013). However, as the allelopathic effects of black locust have not yet been tested under natural conditions, it seems that vegetation change towards ruderal and nitrophilous species is caused by the change in the availability of soil nutrients rather than by allelopathy (Vítková and Kolbek 2010).
Once introduced in an area, black locusts expands readily into open sites where they form dense colonies and their shade reduces competition from other sun-loving species hindering their natural regeneration and thus a local, small-scale reduction of biodiversity can occur (Kutnar and Kobler 2013).
However, black locust has also some positive effects on biodiversity (Vítková et al. 2016: (a) Provides habitat for some rare and endangered plant species (e.g., rare ruderal plants). (b) In species-poor agricultural landscapes, black locust stands can increase diversity and play a significant role as migration corridors for woodland animals. (c) The dense undergrowth of shrubs and complex canopy structure provide nesting opportunities and a food source for many bird species. Among the birds are habitat generalists rather than specialist species, and their total number can be higher in the stands invaded by black locust than in, for example, oak stands. (d) Old black locust individuals can provide a refuge for saprophagous beetles, including rare species.
In native range, black locust is considered an invasive in the prairie and savanna regions of the American Midwest where it can dominate and shade these open habitats (Swearingen and Bargeron 2016).
The invasion by root suckers occurs only around areas where it is already established, i.e., invasiveness is limited and the threats that the species could pose to biodiversity are localized. As mentioned, black locust does not occur in dense forests or shrublands, with closed canopies: as seedlings are intolerant of shade, mortality is high in closed forests (Boring and Swank 1984;Kowarik 2011;Boer 2013;Crosti et al. 2016;Vítková et al. 2016Vítková et al. , 2017. In such stands, it can only occur as individual trees or groups of trees after disturbance results in a gap in the canopy (e.g., after trees die, fire, windthrow or forest management) where there is less competition (Cierjacks et al. 2013;Vítková et al. 2017).
In general, habitats that are resistant to black locust invasion (Crosti et al. 2016;Vítková et al. 2016) include: (1) areas with compacted and poorly aerated soils such as gleysols in alluvial forests subject to frequent and long-term waterlogging; (2) intensively managed sites where the vegetation is permanently disturbed (e.g., trampled sites, mown lawns and meadows or arable fields); (3) dense forests of competitive shade-tolerant species such as European beech (Fagus sylvatica L.); and (4) mountain and subalpine areas with frosts, short growing seasons and cold climate. These are the reasons why black locust is sometimes considered only a "partly invasive" species as proposed by German forest experts .
Interestingly, agricultural landowners consider black locust as invasive only when it spreads onto agricultural lands; they appreciate and manage black locust stands on poor and dry sites (e.g., sand dunes, south-facing slopes, with shallow soils), where it performs better than other native tree species (Kutnar and Kobler 2013;Stančič 2015).
The current range of black locust in Europe is under strong temperature constraints, and it is projected to expand in response to climate change (Boer 2013;Kutnar and Kobler 2013;Giuliani et al. 2015;Dyderski et al. 2017), so the species is expected to invade new sites and also increase its population in already occupied areas (Kleinbauer et al. 2010). It is the case of managed temperate forests of Europe, currently dominated by European beech and Norway spruce (Picea abies (L.) Karst), where species like black locust, northern red oak (Quercus rubra L.), Turkey oak (Quercus cerris L.) and fluttering elm (Ulmus laevis Pall.) are considered as interesting alternatives if the predicted northward shift of 461 km and 697 km for thermophilic species under warming scenarios of 2.9 °C and 4.5 °C, respectively, over the 2060-2080 period is realized (Thurm et al. 2018).
Under climate changes, preventive and control measures are needed and should include: (1) Establishment of a managed buffer zone around black locust plantations to act as a biological barrier for the species (Crosti et al. 2016); (2) Planning its introduction close to and within areas of ecological and environmental value (e.g., Natura 2000 sites) as well as in rare or threatened forest habitats (Borde 2011;Crosti et al. 2016); (3) Conversion from coppice to high forests by cessation of coppice cuts combined with lengthening rotation age and use of an alternative cutting system (Radtke et al. 2013;Nadal-Sala et al. 2019; La Porta pers.comm.); (4) Application of mechanical, chemical or biological control measures to eradicate or reduce the presence of black locust.
The detailed application of these preventive and control measures will be presented in the 'Management of black locust' section of this article.
However, we fully agree with Vítková et al. (2016Vítková et al. ( , 2017 and Sádlo et al. (2017) that a stratified and sitespecific management approach, tolerating (or even favouring) black locust in selected areas and strictly eradicating it from naturally valuable habitats, is the only solution for a strict and rational management of the species in Europe. This approach should take into account both the ecological and economic aspects associated with the occurrence of black locust in different types of Robinia habitats.

Shade tolerance
In both the native range (Trimble 1975, cited by Huntley 1990DeGomez and Wagner 2001) and in Europe (Poskin 1926;Negulescu and Săvulescu 1957;Haralamb 1967;Herman 1971;CRPF 2007;Brus 2011;Vuković et al. 2013), black locust is considered a strong lightdemanding tree species, intolerant of shade (growing best in full sun) and competition; therefore it is found in closed stands only as a dominant tree (McAlister 1971;Huntley 1990), requiring spacing at crown level for optimal growth (Poskin 1926). Black locust grown on poor sites cannot withstand shade so self-thinning is very high under such conditions (Herman 1971). Lyr et al. (1963) found a lower shade tolerance of black locust compared to Q. rubra, Pseudotsuga menziesii (Mirbel.) Franco, Pinus sylvestris L., Betula pendula Roth, Alnus glutinosa (L.) Gaertn. in terms of lower relative growth rates under shade and higher shoot/root ratios under increasing shade. As with other light-demanding tree species (e.g., Populus nigra L., Fraxinus pennsylvanica Marsh.), mature black locust trees show a pronounced phototropism and develop a leaning stem (a deformity reducing the milling potential) in search for light (Evans 1984;MAPPM 2000b).

Potential for natural pruning
Black locust, as a strong light-demanding tree species, shows a good potential for natural pruning but when trees are grown in abundant light, they do not self prune and tend to develop large and deep crowns (Rédei 2013b). Kalmukov (1995) showed the influence of six different planting spacings (1.5 × 0.5 m, 1.5 × 1.0 m, 1.5 × 1.5 m, 2 × 0.5 m, 2 × 1 m, and 2 × 1.5 m) on natural pruning of black locust: at age seven, the branch-free part of the bole ranged from 1.5 m in the widest planting scheme to 2.8 m in the closest one.
However, even when grown in dense stands, it is not a perfect self-pruner so the removal of low branches is necessary to produce 1st grade sawlogs designated for top-quality wood products such as veneer (Haralamb 1967;Rédei 2013b;Meyer-Münzer et al. 2015).
Black locust seeds have been affected in Bulgaria, Romania and Slovakia by insects such as locust sawfly, Eurytoma caraganae Nikolskaya being the most common, as well as lima bean pod borer Etiella zinckenella Tr. and locust seed beetle Kitorrhinus quadriplagiatus Mots., the proportion of damaged seeds reaching ca. 80% (Iliev et al. 2005).
Black locust is considered a species very resistant to wood decay caused by fungi (Glavaš 1999). However, some fungi genera such as Fusarium, Polyporus, Perenniporia, Ganoderma, Laetiporus, and Fomes have caused white, brown or red rot, especially after stems or roots have been damaged (Schütt 1994;Tomiczek et al. 2008;Meyer-Münzer et al. 2015). Black locust roots are also affected by Armillaria spp., causing root decay especially if the trees are stressed by drought (Aslam 2015).
Fortunately, the most important pest in native range, the locust borer Megacyllene robiniae (Forst.), is not yet present in Europe. Tunnels produced by its larvae in the heartwood weaken tree limbs, making them susceptible to breakage by wind and also serve as primary infection sites for windborne spores of the fungus Phellinus rimosus (Berk.) Pilát which causes damaging heart rot disease. This combined attack makes growing black locust for timber production impractical where this pest is present (Huntley 1990).
In countries like Bulgaria, Germany, Hungary, Slovakia, Austria and Romania, the European hare (Lepus europaeus Pallas) and European rabbit (Oryctolagus cuniculus L.) peel-off the bark of young trees (in nurseries and in forest stands), while red deer (Cervus elaphus L.) and roe deer (Capreolus capreolus L.) browse the young shoots, leaves and buds, causing significant damage (Juhásová and Hrubík 1984;Trenchev et al. 1993;Iby 1997;Claessens et al. 2006;Rédei et al. 2012). Similarly, in the USA, white-tailed deer (Odocoileus virginianus Zimmerman) is a major problem for young black locust, with reports of up to 92% of shoots browsed in the southern Appalachians (Della-Bianca and Johnson 1965, cited by Huntley 1990).

Growth and yield dynamics Height growth
In both the native range (Harlow et al. 1986) and in Europe (France, Merzeau et al. 2008;Hungary, Rédei 2013b;Poland, Zajączkowski and Wojda 2012;Romania, Haralamb 1967;Germany, Meyer-Münzer et al. 2015;Carl 2018), black locust grows rapidly in height for the first 10-15 years. Consequently, the trees can reach heights of 10 m at 5 years and 14 m at 10 years on the best sites (Rédei 1984;Ruhm 2013). The peak growth rate in height occurs on the best sites with warm climate, rich and deep soils within the first decade: in the first 5 years (France, Merzeau et al. 2008;Germany, Lockow and Lockow 2015;Hungary, Rédei 1984;Poland, Zajączkowski and Wojda 2012) or in the first 10 years (Austria, Ruhm 2013; Ukraine, Lavnyy pers. comm.; Romania, Negulescu and Săvulescu 1957). The better the site potential, the earlier the peak of height growth. This slows after 15 years (Haralamb 1967), although growth remains relatively rapid in older trees; black locust stands in Europe can reach mean or dominant heights of over 20 m at 20 years and up to 25-30 m at 30 years (Tables 3, 4).
Currently, the tallest trees in Europe are found in France (38 m, IGN 2018) and Germany (39 m, Lockow and Lockow 2015).

Diameter growth
Black locust trees have rapid early diameter growth that can reach 3-4 cm in the first year (stool shoots and root suckers, Jovanović 1967) on the best sites. The diameter increment peaks early on such sites: before age 10 (at 5 years in Germany, Lockow and Lockow 2015; 6-8 years in Poland, Węgorek and Kraszkiewicz 2005;Klisz et al. 2014;7 years in Slovenia, Kadunc 2016), at the end of the first decade (Austria, Ruhm 2013;Bulgaria, Dimitrov 2001;Hungary, Rédei 1984;Rédei et al. 2008;France, Merzeau et al. 2008) or between 12 and 20 years (Romania, Negulescu and Săvulescu 1957;Belgium, Claessens et al. 2006;Slovenia, Kadunc 2016). The better the site potential, the earlier the peak of diameter growth (e.g., from 7 years on the best sites to 18 years on the poorest in Slovenia, Kadunc 2016).
As in the case of height, stool shoots have larger diameters than root suckers and planted seedlings up to 15-20 years of age. The trend in diameter growth rate reverses later with the diameter of seed-originating stands exceeding that of coppice stands of the same age (Haralamb 1967).
In the best growing conditions in native range and in Europe, 20 cm diameter trees take 25-30 years to be produced (Roach 1965, in Huntley 1990Giurgiu and Drăghiciu 2004;Kalmukov 2014;Rédei et al. 2014). Trees 50 cm in diameter can be produced before 60 years of age (Claessens et al. 2006;Dancart 2019).
As black locust grows rapidly in the first decades, wood production can reach important levels quite early: over 200 m 3 ha −1 at 20 years, 290 m 3 ha −1 at 30 years and 350 m 3 ha −1 at 40 years in the best site conditions (Table 6).
However, these data do not reveal the relationship between productivity and method of reproduction; for both stool shoots and root suckers, wood production decreases significantly over time (particularly in the case of stool shoots), from the first to the third generation (Table 7).

Management of black locust Goals
The goals of managing black locust depend on the location of the stand (e.g., areas of high ecological and environmental values versus stands under regular forest management) and the function (e.g., protection/conservation vs. wood or biomass production) ).
In areas with high ecological and environmental values, such as Natura 2000 sites, or with rare or threatened forest habitats, the goal (e.g., Germany, Poland, Slovakia, Ukraine, Vor pers.comm.; Klisz pers.comm.; Benčať pers. comm.; Lavnyy pers.comm.) is to strictly eradicate existing black locust from species compositions and return to native species (Vítková et al. 2016). To avoid further invasion of black locust into as yet un-colonized spaces within protected areas, the prevention of introducing the species by deliberate planting, based on legal restrictions, is the target (Kleinbauer et al. 2010;Boer 2013).

Management of black locust in areas with high conservation values
In such areas, the main goal is strict eradication and replacement with native species. The most important eradication techniques for black locust control are manual, mechanical, chemical, mechanical/chemical and biological.
Manual control consisting of hand pulling is preferred for seedlings and small saplings and should guarantee that no roots are left in the ground (Boer 2013; https ://invas oras.pt/ en/galle ry/robin ia-pseud oacac ia-en).
The mechanical control of black locust includes low cutting and girdling of trees (Sabo 2000;Boer 2013). The most efficient method is incomplete girdling which includes the removal of a strip of bark and sapwood 8 cm wide from 9/10ths of the girth, followed by complete girdling in the following growing season and later low cutting. The tree does not produce further root suckers and stool shoots (Starfinger and Kowarik 2003). However, the efficiency of both low cutting and girdling is debated (Boer 2013) since killing the main stem is often followed by formation of shoots from the tree base (Sabo 2000).
Chemical control includes foliar spraying (late in the growing season) and tree injection (after making incisions into the sapwood). Active chemical substances to be used are glyphosate, triclopyr, picloram, each diluted with water, as well as an atrazine-simazine mixture (Boer 2013;invasoras. pt/en/gallery/robinia-pseudoacacia-en).
Biological control makes use of domestic goats and cattle which defoliate black locust; in the native range after 4 years of browsing, all trees had been killed (Stone 2009).
As complete eradication of black locust is not possible since such treatment is expensive because of continuous stump sprouting, one way to fulfill the objective of replacing the species is prolonging the rotation age until the trees become senescent and are replaced naturally by native species (Motta et al. 2009). Prevention measures, like avoidance of disturbances (e.g., soil disruption) that favor black locust colonization/dispersal by root suckers, are also possible (Boer 2013).

Management in areas with wood production objectives
The rotation length is shortest when black locust is grown for biomass energy production: 3-5 years in very short rotation coppice (Borde 2011;Rédei 2013a, b, c, d;Steinegger and Reh 2013), and is 5-7 years in Italy (Facciotto et al. 2008) or 10 years in France (Bousse and Richard 2017) in short rotation coppice.
In black locust stands for wood production, the rotation age depends on: 1. Type of wood product: rotation age is 50-60 (even 70) years when sawlog (dbh > 40 cm) production is targeted as in Germany (Engel et al. 2014) and Austria (Hochbichler et al. 2015). In France, the rotation age for sawlog production is 40 years and decreases to 25 years for the production of vineyard poles (coppice stands) (Pagès 1986; Bousse and Richard 2017). 2. Production (fertility) class (and correlated type of wood product): it ranges between 20 and 25 years in the lowest production classes (V in Romania, VI in Hungary, for poles, props, pulp, other small-size industrial wood) and 35-40 years in the Ist class for good quality sawlogs (Rédei 1984;MAPPM 2000d). 3. Regeneration pathway this is 20-25 years in coppice stands compared with stands of seed origin (30-35 yearold plantations), as in Bulgaria (Dimitrov 2001). The rotation age in these coppice stands is similar to ones in Germany and Austria (

Silviculture of black locust stands for wood production Natural regeneration
As mentioned earlier, natural regeneration of black locust is primarily by root suckers. The species is also regenerated, on a much smaller scale, by stool shoots and by seed.
When the rotation age has been reached, the main silvicultural system used in black locust stands in Hungary, Romania, Bulgaria, Croatia, France, Germany, Poland, and Slovenia is simple (low) coppice, considered the most cost-effective management system (Sitzia et al. 2016). This creates light-abundant sites suitable for the vegetative (and generative) regeneration of black locust (Radtke et al. 2013). After the cut, the stools are (1) left to produce adventitious or proventitious sprouts, or (2) dug out, followed by ploughing/scarification to wound shallow roots to promote root suckers (Romania, MAPPM 2000c; Bulgaria, Kostov and Alexandrov 2018; Germany, Meyer-Münzer 2015; Ukraine, Lavnyy pers.comm.) (Fig. 1).
This method is cheap and efficient, allowing local people (e.g., from Romania) to collect the stool wood, which is highly valued as firewood. After the third coppice cut at about 90 years of age, after which the sprouting potential-and so the economic viability-decreases (e.g. Table 7), and the proportion of heart rot increases, all black locust stools are removed and the stand is regenerated by planting (Iby 1997;MAPPM 2000c).

Artificial regeneration
Black locust may be established by planting using reproductive material collected in certified/selected seed stands and seed orchards. These exist in Hungary, Bulgaria, Romania, Germany, Italy, Belgium, Czech Republic, Slovakia, Croatia, Bosnia and Herzegovina, FYR of Macedonia, Poland, and Ukraine (Cvjetkovic pers.comm.; Đodan pers.comm.; MMP-RNP-ICAS 2012; Executive Forest Agency 2014; http:// ec.europ a.eu/forem atis; Lavnyy pers.comm.; Mandžukovski pers.comm.). When no local seed sources exist (for example in the case of France where establishment of a seed orchard is in progress, or in Austria), seeds are imported, mostly from other European countries (e.g., Hungary, Romania, Bulgaria), rather than from the United States (Bastien pers. comm.; Pötzelsberger pers.comm.).
The seeds are light and the yield from 100 kg of pods is 20 kg (Haralamb 1967;Rédei et al. 2001;Rédei 2013a). Because the heavy, thick and impermeable seed coat restricts germination, the seeds need pre-treatment to remove seed dormancy. In both native range (Olson 1974;Harlow et al. 1986) and in Europe (Bîrlănescu and Belu 1968;Costea et al. 1969;Savill 2013;Jastrzębowski et al. 2017;Pedrol et al. 2017), mechanical scarification using rotary tools or electric grinders, soaking in concentrated or diluted sulphuric acid, soaking in boiling or nearly boiling (70-80 °C ) water, using low liquid nitrogen or high temperatures in an incubator are the usual methods. After appropriate pre-sowing treatment, germination rates can be as high as 85.0 ± 5.8% (Kheloufi et al. 2018) or 89.0-100% (Cierjacks et al. 2013;Vuković et al. 2013).
Seeds are sown in bare-root nurseries from late April to early May to avoid late frosts. The minimum distance between rows is 35-40 cm (Costea et al. 1969;Keresztesi 1988) and 2-3 g of seeds are used per m 2 with a sowing depth of 2-4 cm (5-6 cm on sandy soils prone to excessive summer warming) (Rubțov 1958;Vakulyuk and 1998;Iancu 1999). The bare-root seedlings are ready for planting in 1 year after reach a minimum 6-mm root collar diameter and 50-200 cm in height (Iancu 1999;Rédei 2013a). In addition, artificial regeneration of black locust using plants originating from root cuttings as well as in vitro micropropagation (tissue culture) is carried out in Hungary, Bulgaria, and Poland for the reproduction of superior individuals or varieties, such as the 'shipmast' locust Robinia pseudoacacia var. rectissima Raber, with a very straight bole, extremely durable wood, and resistance to fungal diseases and insect pests, and Robinia pseudoacacia f. semperflorens (Carr) Voss. (up to five blooming periods during a vegetation season) (Keresztesi 1988;Kalmukov 2000Kalmukov , 2012Szyp-Borowska et al. 2016).
The initial stocking density of black locust plantations for timber production in Europe is highly variable and ranges from 1100 to over 5000 plants ha −1 ( Table 8).
The distance between rows should be wider than that between the seedlings in the row, approximately 2.5-3.0 m to allow for mechanical cultivation (Austria, Ruhm 2013; Bulgaria, Ministry of Agriculture and Food 2013; France, French Ministry of Agriculture 2018; Croatia, Đodan pers. comm.).
Protection, either for individual trees or stands using tubes, grids, or fencing (a minimum height of 1.5 m), is recommended in some countries (Austria, Bulgaria, France, Germany), and is necessary when regenerating black locust   (photo V.N. Nicolescu) in areas with high browsing pressure (e.g., hares, roe deer, red deer as well as domestic goats or cattle) (Iby 1997;Ministry of Agriculture and Food 2013;Meyer-Münzer et al. 2015;Lemonnier Nursery 2018). After 2-3 coppice cycles, growth depression with a reduction of timber production and an increase of heart rot have been observed (Berthold et al. 2005;Borde 2011;Meyer-Münzer et al. 2015;Vítková et al. 2016). Standard practice in some European countries (e.g. Austria, Romania, Bulgaria, Germany) is for planted trees to be coppiced three times, after which stools are removed and the stand regenerated by planting (Iby 1997;MAPPM 2000c;Dimitrov 2001;Meyer-Münzer et al. 2015). In France, black locust coppice stands are managed with four or even five coppice cycles (Borde 2011). However, when the state of coppice stands is poor, they may be clearcut and replanted after only one or two coppice cycles in order to restore the production potential.

Young stand management
Widely-spaced plantations do not require any release cuttings but one-two interventions, on a cycle of 1-3 years are necessary in stands vegetatively regenerated from root suckers or stool shoots, targeting the removal of badly shaped or forked trees ('wolves') as well as to favour root suckers over stump shoots (Haralamb 1967;Debenne 1988;MAPPM 2000b;Meyer-Münzer et al. 2015). However, such interventions are costly, particularly as the wood removed is not marketable. Therefore, combined with the high natural mortality in young root sucker-originating stands, which can be up to 70% in the first 5 years (Pagès 1985, in Carbonnière et al. 2007), these interventions are not recommended in countries like France (Carbonnière et al. 2007;Borde 2011).
At ages 4-5, the first cleaning-respacing (mostly negative selection) is performed to remove defective stems (i.e., forked, badly shaped, wounded, or bent-over as an effect of strong phototropism), to reduce the number of shoots per stool, adjust the spacing between root suckers and reduce stocking to a maximum of 2500 stems ha −1 (Zajączkowski 2013;Wojda et al. 2015). After 3-4 years, this is followed by second cleaning-respacing, mostly based on positive selection, with the same aims but especially to reduce stocking to ca. 1700 or even 2000 stems ha −1 , with spacing as regular as possible to provide additional growing space to the remaining individuals (Constantinescu 1976;Halupa and Rédei 1988;Rédei et al. 2011;Wojda et al. 2015).
A more intensive approach is practiced in France, with much lower initial stocking of plantations (Table 8) which is reduced, after the second cleaning-respacing, to 1000-1200 stems ha −1 . In addition, extraction racks of 3-4 m wide, and at 10 m intervals are considered necessary due to the high stocking of young stands. These are opened during the first cleaning-respacing operation (Debenne 1988;Carbonnière et al. 2007;CRPF 2007).
The wood volume removed by cleaning-respacing may be 4-6 m 3 ha −1 (first intervention) and 15-20 m 3 ha −1 (second intervention), depending on the site potential and productivity class (Rédei 1984). The correct application and timing of cleaning-respacing is the key to good management (Halupa and Rédei 1988). If this intervention is too late or too light, the remaining trees do not develop their crowns normally and they are deformed or too small, as black locust is a strong light-demanding species. The crowns do not recover no matter how much light is subsequently available (Haralamb 1967;MAPPM 2000b).

Commercial thinning
In black locust stands, early, selective, moderate and frequent thinning is undertaken. The operations are carried out at 5-7 years intervals and are either mixed as in Romania (Costea et al. 1963;MAPPM 2000b) or from above as in Austria (Hochbichler et al. 2015), Germany (Vor pers. comm.), France (Borde 2018) and Slovenia (Čarni et al. 2016). Thinning aims to keep the crowns of the final crop trees (e.g., those which are cylindrical, with straight grain, monopodial growth, and healthy dense foliage) free from competition to encourage vigorous growth to reach the target diameter as soon as possible (Borde 2018;Vor pers.comm.). During thinning from above, 2-3 (CRPF Bourgogne 2014), 2-4 (Hochbichler et al. 2015) or a minimum of 4-5 (Borde 2018) of the strongest competitors are removed for each final crop tree. It is advisable to keep the viable trees in the lower (IV and K) Kraft classes as these contribute to the natural pruning of dominant black locust trees and reduce the danger of epicormics by shading the trunks of the most valuable individuals in the upper classes (Bîrlănescu and Clonaru 1960).
The intensity of each thinning intervention is generally between 15 and 25% of the standing volume, although the volume removed per intervention varies between different European countries and ranges from 18 to 30 m 3 ha −1 in Croatia (Đodan pers.comm.), 20 m 3 ha −1 in Serbia (Andrašev pers.comm.), 30-50 m 3 ha −1 in Hungary (Rédei 1984) to 80 m 3 ha −1 in Germany (Lockow and Lockow 2015). In countries like Hungary and Bulgaria, 25-35% of the total wood production is removed during thinning operations (Rédei 2013c). In Hungary, the thinning intensity is determined on the basis of the "growing space index", calculated as the mean distance between trees × 100/mean stand height after thinning. This should be 23-24% in pure black locust stands (Rédei 1984;Rédei 2013c).
Generally, the number of trees per ha after the last thinning equals the stocking at rotation age and ranges between 120 and 150 stems in Austria (Hochbichler et al. 2015) and 700-800 stems in Italy (Maltoni et al. 2012), depending on the country and intended range of final products (Table 9).

Pruning
The terminal shoot of black locust may be repeatedly affected by frosts, causing forking multiple. Rectifying this by formative pruning (shaping) is an intervention-to-waste so is not advisable.
Green high pruning is tolerated well by black locust (Halupa and Rédei 1988) and is carried out only on crop trees, from 120 to 160 individuals ha −1 in Austria (Pötzelsberger pers.comm.) to 300/400-600 individuals ha −1 in France (Merzeau et al. 2008;CRPF 2011). This is completed in two or three lifts and stems are cleared of branches to a height of 4-6/7 m (Halupa and Rédei 1988;Rédei et al. 2017; CRPF Aquitaine 2018) or even 6-8 m (Dancart 2019). The crown length after pruning should be at least a third of the tree height (Meyer-Münzer et al. 2015).

Conclusions
Black locust is a light-demanding, competition-intolerant, fast-growing and relatively tall tree species, able to adapt to difficult site conditions (e.g., dry climates, with nutrient poor and dry soils). It is an important multipurpose species with major economic, ecological and social roles for producing wood, fodder, honey, a source of bio-oil, for biomass production and carbon sequestration, soil stabilization/ erosion control, re-vegetation of landfills, mining areas and wastelands, in biotherapy, and in landscaping. It grows rapidly when young, and height, diameter and volume growth peak before the age of 20 years. It is wind-firm, not affected by any major pest or disease, and coppices easily. Consequently black locust has a simple silvicultural management regime, clearcutting with regeneration either naturally from root suckers or stool shoots, or by planting, with the aim of producing valuble wood with various end-uses, from parquet, veneer, furniture, barrel staves, vineyard stakes to firewood, under short rotation periods of up to 40-50 years compared to major European native broadleaved tree species such as oaks and European beech.
As its potential to adapt to predicted climate changes, particularly drought, seems to be high, the importance of black locust, a species both "beloved and despised"  in Europe, is expected to increase in the future. However, not only will its importance increase under expected climate conditions, but also its invasiveness potential. This reality necessitates a more complex, integrated approach to the management of such controversial species, i.e., largescale use or tolerance in forest stands in areas of economic importance, and strict eradication in other, ecologically valuable or sensitive sites. Acknowledgements This paper is an output from the European COST Action FP1403 'Non-native tree species for European forests-experiences, risks and opportunities' (NNEXT) which was active 2014-2018. The authors also acknowledge helpful suggestions from the reviewers and the editor.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat iveco mmons .org/licen ses/by/4.0/.   (2015) Sawlog production 400-500 (600) Hungary Rédei et al. (2017) Sawlog production 400-700 Poland Wąsik pers.comm. 500