Observations on two non-native Alder species (Betulaceae) naturalising in Ireland

Non-native Alnus species are widely planted in Ireland for landscaping purposes, such as roadside plantings, amenity plantings and revegetation of quarries. This paper gives details of nine locations in Ireland where natural regeneration of A. cordata and A. rubra has been observed. The potential impacts on native habitats, should these species naturalise more widely, are discussed.


Introduction
Alnus glutinosa (Alder) is the only native member of the Genus Alnus in Ireland (Parnell and Curtis, 2012). It is widespread and occurs naturally on a range of sites but with a preference for wet and fertile sites, including sheltered mountain sides, lake shores and river banks, wet sandy soils and gravels (Fennessy, 2004). A. glutinosa is increasingly being planted in commercial forestry due to its fast growth rates and breeding programmes are in place to grow so-called "plus-trees" to produce trees suitable for timber production (Teagasc, 2019). It is also an important component of the Irish government's native woodland afforestation and conservation programme, particularly in riparian areas (Little et al., 2008).
A number of non-native Alnus species have been planted in Ireland, principally for landscaping purposes, such as roadside plantings, amenity plantings and revegetation of quarries. There has recently been increasing interest in the use of non-native Alnus species in Britain for short rotation biomass production or as nursery species for other commercial species due to their ability to fix nitrogen into soil via colonies of the bacteria Frankia alni living in root nodules (Wilson et al., 2018). Planting trials of non-native Alnus also exist in Ireland for their forestry potential (McCarthy, 1979;Wilson et al., 2018). Alnus species produce huge volumes of seeds in cones that are dispersed by wind and water, some species also spread via clonal suckering. This makes them strong candidates for naturalising where suitable site conditions allow. Parnell and Curtis (2012) only list A. incana (Grey Alder) as a naturalised Alnus species in Ireland where it has mainly been observed spreading via suckering.
This paper details locations in Ireland where two other non-native Alnus species, A. cordata (Italian Alder) and A. rubra (Red Alder) have been observed by the author to be naturalising. Where useful, the land history of the site is described based on historical aerial photos available via the Ordnance Survey Ireland Online Mapviewer (http://map.geohive.ie/mapviewer.html). Potential ecological impacts on native ecosystems, should these species become more widespread in the wild, are discussed.

Alnus cordata
This European species (Fig. 1) has a very limited native range, being confined to the southern Apennine mountains in Italy and north-eastern mountains of Corsica where it is a pioneer tree species of Fagus sylvatica and Castanea sativa woods, growing as pure stands in gaps (Ducci & Tani, 2009). Unlike native A. glutinosa, A. cordata can grow in much drier conditions (Wilson et al., 2018). A. cordata is frequently planted in parks, streets and shelter belts (Johnson, 2004).  2.9 ha of an abandoned building site located in an urban setting, formerly pasture and hedgerows, based on historical aerial photos from the early 2000s, but now consisting of bare and re-colonising ground. A high density of saplings was observed regenerating at the edges of the site on bare ground with a scattering of saplings in the centre. Seed source is from trees planted for landscaping in surrounding housing estates and the adjacent Kinsale Road.

5) v.c.H27 West Mayo
Townland: Garryduff; M 14295 90406; year observed:2019. 0.07 ha of rank wet grassland on the north bank of the Castlebar River in an urban environment. A dense stand of A. cordata saplings was observed regenerating in the grassland. The seed source was adjacent mature trees planted on the roadside. A. cordata has been extensively planted in Castlebar, particularly as plantations along the N5 road.

Alnus rubra
One of the largest Alnus species in the world, this North American species is native to the Pacific Northwest (Johnson, 2004). It is principally found in forests below 750 m ASL within 200 km of the coast where it is a pioneer and riparian species that grows alongside Picea sitchensis, Thuja plicata, Aibies grandis, Pseudotsuga menziesi and Tsuga heterophylla (Wilson et al., 2018), conifers that dominate commercial forestry in Britain and Ireland. It is planted much less frequently than A. cordata, usually as a specimen tree or as a shelterbelt (Johnson, 2004). However, in Southwestern Ireland it has also been planted in commercial forestry plantations in riparian areas or edges (see below).
A. rubra saplings are growing on an overgrown forestry track, along with Betula pubescens and S. cinerea in a forestry plantation (Fig. 3). The seed source for regeneration on the track was a single mature tree growing next to the barrier gate, which may itself have originated from seed blown from mature trees growing in a nearby garden centre.
This area is predominantly commercial forestry plantations and A. rubra saplings are growing on the edges of forestry roads. The seed source is mature stands planted along the roads and plantation edges.
There are mature roadside plantings of A. rubra on both sides of the N22 at this location, planted when the road was upgraded in the 1980s and 1990s. A small amount of natural regeneration was observed on the roadside verges. This location is heavily afforested so there is potential for roadside stands to seed into clear-fell sites and forestry tracks.
Natural regeneration of A. rubra was patchy but quite dense in the wetter parts of the site (Fig. 4). Natural regeneration of A. glutinosa and B. pendula was mixed in with A. rubra. Seed source was mature roadside plantings on the southern property boundary.

Discussion
The locations where A. cordata and A. rubra have been observed to be naturalising are sites that have been highly modified by human activities i.e. urban waste ground, roadsides, forestry plantations. Urban brownfield sites typically host a number of alien pioneer and later successional species, with species composition determined by history of the site, soil structure and composition and seed sources in the surrounding environment (Lososová et al., 2016). They often offer a unique opportunity to study how plant species from vastly different parts of the world interact when growing together in what have become known as "novel ecosystems" (Hobbs et al., 2009). The conservation value of these ecosystems is currently being hotly debated in the field of conservation biology. The short lived nature of brownfield sites, such as the one found with A. cordata and B. davidii on the River Tolka in Dublin often mean that it is not possible to study how these sites develop in the long term, particularly sites with emergent woodland communities. As brownfield sites are generally isolated blocks of habitat, they offer an interesting opportunity to study interactions of alien species and native species without a high risk of invasive species spreading. What would the impact on native semi-natural habitats be should A. cordata and A. rubra become more widespread as naturalised species? Both species act as pioneer woodland species in their native ecosystems so would likely be easily replaced by later successional native or non-native tree species. These species could compete with native pioneer tree species, such as B. pubescens, B. pendula, Salix spp. but in particular, the native A. glutinosa. The amount of niche overlap between the native and non-native Alnus species would need to be determined but the natural range of A. cordata and A. glutinosa overlap in Corsica and southern Italy (Kajba & Gračan, 2003;Ducci & Tani, 2009), which might suggest niche partitioning between these species. The ability of A. cordata to grow in much drier site conditions than A. glutinosa lends credence to this. There may be greater overlap between A. rubra and A. glutinosa, both of which are principally found in wet soils in their native ranges (Kajba & Gračan, 2003;Wilson et al., 2018). Being a much larger tree, A. rubra seedlings could out-compete A. glutinosa seedlings through direct competition for space. A. rubra, however, does not seem to do well in areas with late frosts and tends to show better growth in Britain in areas with an oceanic climate (Forest Research, 2018;Savill, 2019) so may only be able to thrive in Ireland along the western seaboard, where the climate is very similar to its native range in the Pacific Northwest. Indeed, it has been proposed that A. rubra could be planted in mixed stands with P. sitchensis in upland plantations in Britain and Ireland in order to enhance their structure and biodiversity potential by recreating the natural temperate rain forests in their native range where both species grow together (Deal et al., 2014).
The genetic integrity of A. glutinosa could also be threatened through hybridisation with non-native Alnus species (Wilson et al., 2018). Hybridisation between A. glutinosa and the non-native A. incana has been recorded in Ireland (Parnell 1994; BSBI Distribution Database, https://database.bsbi.org/). Hybridisation however, is thought to be uncommon in other parts of Europe where both species coexist, due to differences in flowering times and low seed production of hybrids (Banaev et al., 2007). A study to determine the presence of hybrids under natural conditions in Ireland at sites where A. glutinosa and A. cordata and/or A. rubra are growing together would be worthwhile. Such a study should look at the potential of hybridisation in these populations based on the overlap of flowering times between these species and the frequency of hybrids in natural regeneration.
The invasiveness of both A. cordata and A. rubra could also possibly be kept in check by the pathogen Phytophthora alni (Bjelke et al., 2016). The ability of Alnus species to fix nitrogen in soil could have impacts on native plant communities in Ireland that require low nitrogen levels and where the native A. glutinosa doesn't naturally occur but where A. cordata or A. rubra could grow, however, neither species does well in peat soils (Wilson et al., 2018). A similar issue has been found with Gunnera tinctoria in western Ireland (Mantoani et al., 2020). Long term studies on naturalised stands of A. cordata and A. rubra would be useful in determining both their invasiveness and their potential to either enhance or hinder the biodiversity of both semi-natural and human-made habitats. Impacts on A. glutinosa could be determined by establishing trials of mixed stands with A. cordata, A. rubra or both, by setting seeds in controlled conditions that replicate those where the native species is found in the wild, in order to study competitiveness. Similar trials could be conducted with other native pioneer trees species. It would also be useful to investigate the value of A. cordata and A. rubra for native wildlife in comparison to native A. glutinosa. The author has observed large roving flocks of Carduelis spinus (siskin) foraging for seeds from cones in mature plantations of A. cordata in Castlebar, Co. Mayo.