Access the full text.
Sign up today, get DeepDyve free for 14 days.
Loading next page...
/lp/ou_press/evidence-from-mortality-dating-of-fraxinus-excelsior-indicates-ash-ogO1ni5SG6
- Publisher
- Oxford University Press
- Copyright
- © Crown copyright 2018.
- ISSN
- 0015-752X
- eISSN
- 1464-3626
- DOI
- 10.1093/forestry/cpx059
- Publisher site
- See Article on Publisher Site
Abstract
Abstract Ash (Fraxinus excelsior L.) planted at six sites over the past 20 years was investigated. Three geographically isolated sites (Northumberland, Leicestershire and Devon) were compared with three sites in established areas of ash dieback in East Anglia, and the causal pathogen, Hymenoscyphus fraxineus, confirmed at all. Dieback severity, the frequency of stem basal lesions and pathogen apothecia, were quantified at all sites but despite high disease levels, tree mortality was low. Some trees had typical H. fraxineus stem cankers but had apparently died between 2001 and 2011, before the earliest UK records of H. fraxineus. Ring counts established beyond doubt the year of death and canker initiation in 27 dead trees. Cankers on the same trees were then tested for H. fraxineus using PCR-based detection, with pathogen presence confirmed as early as 2004/05 in some. This places H. fraxineus in England much earlier than previously thought, even pre-dating its documented arrival in neighbouring European countries. The advanced disease levels at some sites plus confirmation of H. fraxineus in old stem cankers, suggests that planting of infected H. fraxineus tree stock in England could date back to the early 1990s, with affected trees dying in the mid-2000s. Additionally, it raises questions about the origins of the infected plants and uncertainties about plant trade pathways. Introduction The fungus Hymenoscyphus fraxineus (T. Kowalski) Baral, Queloz & Hosoya (basionym: Chalara fraxinea, synonym: H. pseudoalbidus) is now acknowledged as the cause of serious dieback (Kowalski and Holdenrieder, 2009a,b) which threatens populations of European ash (Fraxinus excelsior L.) and narrow-leaved ash (F. angustifolia Vahl) across much of Europe (Keßler et al., 2012). The first indication of the current epidemic in Europe dates from observations in the early 1990s of severe dieback of ash in Poland (Przybyl, 2002) and Lithuania (Juodvalkis and Vasiliauskas, 2002). However, at that time frost and drought were implicated as the cause of the symptoms and a biotic causal agent was not identified and named until about a decade later (Kowalski, 2001, 2006). It is now clear that H. fraxineus is an introduced pathogen which appears to originate from East Asia, with reports of the fungus from China, Japan, Korea and Russia (Zhao et al., 2013; Baral and Bemmann, 2014; Han et al., 2014; Zheng and Zhuang, 2014). In its native range associated with Asian ash species (e.g. F. mandshurica Rupr. in Japan) H. fraxineus is reported to cause little or no damage (Zhao et al., 2013; Drenkhan et al., 2017). However, the behaviour of the fungus on native European ash species (F. excelsior and F. angustifolia) is very different (Keßler et al., 2012). Initially disease symptoms on trees include foliage necrosis but infection then progresses from affected leaves and rachises (petioles) into shoot or stem tissue, killing the cambium and phloem and forming characteristic diamond-shaped lesions which can completely girdle affected branches and stems. To complete its life cycle H. fraxineus then sporulates by producing fruit bodies (apothecia) during summer months formed mainly on the rachises of fallen leaves infected the year before but also occasionally on infected dead shoots, stems and root collars of young ash trees in ground contact (Gross et al., 2012; Kirisits et al., 2012). However, apothecia can only be produced if both mating types (MAT1-1 and MAT1-2) of the fungus are present and sexual reproduction occurs (Gross et al., 2012). Once apothecia are formed in any number the disease can start to build-up to epidemic phase as wind-dispersed ascospores are released (Kowalski and Holdenrieder, 2009b; Timmermann et al., 2011), leading to multiple infections on the foliage of nearby ash trees as dispersal of the spores can occur over hundreds of metres (Chandelier et al., 2014). Hymenoscyphus fraxineus was first identified in February 2012 in southern England on nursery stock that had been imported from the Netherlands (Sansford, 2013). It was subsequently confirmed in June 2012 on a landscaped site planted the previous year in Leicestershire (DR Rose and JF Webber, Forest Research, pers. comm.). Following this, the Forestry Commission (FC), Food and Environment Research Agency and other Government departments initiated an extensive surveillance programme, surveying stock in nurseries, new plantings and ash growing in the wider environment. As a result, more findings were made including a small number of trees on sites where ash had been planted up to five years earlier, providing evidence that some trees had been infected prior to planting out (Sansford, 2013). By October 2012 H. fraxineus infection had been found at many locations in the wider environment, particularly across East Anglia and Kent, with naturally regenerating saplings in established woodland and hedgerows showing symptoms. These wider environment findings indicated the early stage of an epidemic was underway, so a systematic survey of ash across the UK was undertaken in November 2012. It identified a total of 184 infected sites; 114 in the wider environment and 55 associated with recent plantings plus 15 nursery sites (Clark and Webber, 2017). At the conclusion of 2012, the distribution of positive findings showed a concentration of wider environment infected sites along the eastern seaboard of Kent, East Anglia and sporadically in Lincolnshire and East Yorkshire northwards to Northumberland and Scotland (Freer-Smith and Webber, 2015). This suggested the possibility of disease establishment as a result of spore inoculum blown in the wind from continental Europe. Modelling undertaken to evaluate this concluded that there were at least 100 days between 2008 and 2011 on which environmental conditions including wind direction, rainfall and humidity, were conducive for dispersal of ascospores from infected ash in mainland Europe across the sea, thereby causing infection of ash predominantly in east and south-east England (PostBOX, 2012). However, there were also more isolated areas of infection with visible H. fraxineus symptoms concentrated within woodlands newly established since 2007, and some of the stands even had planting dates well before the disease had been recognized in Europe. With sites planted in the 1990s up to the mid-2000s, especially those located in more isolated areas, infection from windblown spores originating from outbreaks in neighbouring European countries appears very unlikely. Therefore, although ash dieback disease may have established in the UK as result of airborne inoculum of H. fraxineus from mainland Europe, disease foci were also likely to have been initiated from infected imported stock in some parts of the country. We set out to explore the likelihood of disease spread from infected planting stock, and to establish how long ago this might have happened on sites which had a documented history of planting and subsequent management. Materials and methods Site selection In 2013, Forestry Commission England Plant Health surveyed ash stands planted from 1991 onwards for the presence of symptoms consistent with H. fraxineus. The survey identified a number of sites that were unlikely to have established from windborne inoculum but where disease appeared to have been present well before the first confirmed findings in 2012. Resources limited the number of sites that could be studied but two categories were defined: (1) Sites located in areas of established and widespread infection (in eastern England) and (2) sites geographically isolated from areas of established and widespread infection. Additional criteria for site selection were: each had a significant component of ash planted in blocks or groups; comparability in age and accessibility; a planting date before 2012; and a geographic distribution which gave national representation. All sites also contained a number of standing dead trees, many of which had cankers indicative of H. fraxineus infection prior to death. In all, six sites were selected and the presence of the pathogen confirmed in trees with current symptoms (see King and Webber (2016) for methods). The sites shown in Figure 1 comprised one in Devon, one on the Derbyshire/Leicestershire border, one in Northumberland and three in East Anglia where ash dieback symptoms were relatively widespread in 2012. Sites selected in East Anglia consisted of two neighbouring sites in Norfolk and one in Suffolk. Further details about each site including origin of the plants and year of planting are supplied in Tables 1 and 2. Figure 1 View largeDownload slide Locations of the six study sites in England. Site details are: (1) Wooler, Northumberland; (2) Bickleigh, Devon; (3) Swadlincote, Derbyshire/Leicestershire border; (4) Framlingham, Suffolk; (5, 6) two sites both near Reepham, Norfolk. Figure 1 View largeDownload slide Locations of the six study sites in England. Site details are: (1) Wooler, Northumberland; (2) Bickleigh, Devon; (3) Swadlincote, Derbyshire/Leicestershire border; (4) Framlingham, Suffolk; (5, 6) two sites both near Reepham, Norfolk. Table 1 Fraxinus excelsior sites in England evaluated for ash dieback with dating of tree mortality. Location Stand Area (ha) Assessed compartment Symptom Frequency1 (in assessed compartment) Assessment of dead trees Plant origin Year of planting2 Basal lesions Dead trees No. of trees assessed Estimated years growth Estimated year of death Site 1 1.2 Nursery in Scotland 1999/2000 No data Scarce 2 trees 3 2001/02 near Wooler, Northumberland Site 2 28 Nursery in Belgium 1996/1997 Common Scarce 1 tree 10 2004/05 near Bickleigh, Devon 2 trees 11 2005/06 1 tree 12 2006/07 1 tree 15 2009/10 Site 3 27 Nursery in England 2001/2002 (estimated) Scarce Scarce 3 trees 8 2007/08 near Swadlincote, Derbyshire / Leicestershire border 2 trees 9 2008/09 1 tree 11 2010/11 1 tree 12 2011/12 Site 4 90 Nursery in England 1991/1992 Occasional Scarce 1 tree 6 1995/96 near Framlingham, Suffolk Site 5 3 Nursery in England 1996/1997 Common Occasional 4 trees 8 2002/03 near Reepham, Norfolk 1 tree 9 2003/04 1 tree 10 2004/05 3 trees 11 2005/06 1 tree 12 2006/07 Site 6 8 Nursery in England 1996/1997 Common Occasional 4 trees 7 2001/02 near Reepham, Norfolk 5 trees 8 2002/03 1 tree 13 2007/08 1 tree 15 2009/10 Location Stand Area (ha) Assessed compartment Symptom Frequency1 (in assessed compartment) Assessment of dead trees Plant origin Year of planting2 Basal lesions Dead trees No. of trees assessed Estimated years growth Estimated year of death Site 1 1.2 Nursery in Scotland 1999/2000 No data Scarce 2 trees 3 2001/02 near Wooler, Northumberland Site 2 28 Nursery in Belgium 1996/1997 Common Scarce 1 tree 10 2004/05 near Bickleigh, Devon 2 trees 11 2005/06 1 tree 12 2006/07 1 tree 15 2009/10 Site 3 27 Nursery in England 2001/2002 (estimated) Scarce Scarce 3 trees 8 2007/08 near Swadlincote, Derbyshire / Leicestershire border 2 trees 9 2008/09 1 tree 11 2010/11 1 tree 12 2011/12 Site 4 90 Nursery in England 1991/1992 Occasional Scarce 1 tree 6 1995/96 near Framlingham, Suffolk Site 5 3 Nursery in England 1996/1997 Common Occasional 4 trees 8 2002/03 near Reepham, Norfolk 1 tree 9 2003/04 1 tree 10 2004/05 3 trees 11 2005/06 1 tree 12 2006/07 Site 6 8 Nursery in England 1996/1997 Common Occasional 4 trees 7 2001/02 near Reepham, Norfolk 5 trees 8 2002/03 1 tree 13 2007/08 1 tree 15 2009/10 1Categories are quantified as: Scarce < 1 per cent, Occasional 1–10 per cent, Common 11–50 per cent. 2After the main planting year some replanting was undertaken the following year to replace plants that failed to establish, so planting and year of death may be +1 year. View Large Table 1 Fraxinus excelsior sites in England evaluated for ash dieback with dating of tree mortality. Location Stand Area (ha) Assessed compartment Symptom Frequency1 (in assessed compartment) Assessment of dead trees Plant origin Year of planting2 Basal lesions Dead trees No. of trees assessed Estimated years growth Estimated year of death Site 1 1.2 Nursery in Scotland 1999/2000 No data Scarce 2 trees 3 2001/02 near Wooler, Northumberland Site 2 28 Nursery in Belgium 1996/1997 Common Scarce 1 tree 10 2004/05 near Bickleigh, Devon 2 trees 11 2005/06 1 tree 12 2006/07 1 tree 15 2009/10 Site 3 27 Nursery in England 2001/2002 (estimated) Scarce Scarce 3 trees 8 2007/08 near Swadlincote, Derbyshire / Leicestershire border 2 trees 9 2008/09 1 tree 11 2010/11 1 tree 12 2011/12 Site 4 90 Nursery in England 1991/1992 Occasional Scarce 1 tree 6 1995/96 near Framlingham, Suffolk Site 5 3 Nursery in England 1996/1997 Common Occasional 4 trees 8 2002/03 near Reepham, Norfolk 1 tree 9 2003/04 1 tree 10 2004/05 3 trees 11 2005/06 1 tree 12 2006/07 Site 6 8 Nursery in England 1996/1997 Common Occasional 4 trees 7 2001/02 near Reepham, Norfolk 5 trees 8 2002/03 1 tree 13 2007/08 1 tree 15 2009/10 Location Stand Area (ha) Assessed compartment Symptom Frequency1 (in assessed compartment) Assessment of dead trees Plant origin Year of planting2 Basal lesions Dead trees No. of trees assessed Estimated years growth Estimated year of death Site 1 1.2 Nursery in Scotland 1999/2000 No data Scarce 2 trees 3 2001/02 near Wooler, Northumberland Site 2 28 Nursery in Belgium 1996/1997 Common Scarce 1 tree 10 2004/05 near Bickleigh, Devon 2 trees 11 2005/06 1 tree 12 2006/07 1 tree 15 2009/10 Site 3 27 Nursery in England 2001/2002 (estimated) Scarce Scarce 3 trees 8 2007/08 near Swadlincote, Derbyshire / Leicestershire border 2 trees 9 2008/09 1 tree 11 2010/11 1 tree 12 2011/12 Site 4 90 Nursery in England 1991/1992 Occasional Scarce 1 tree 6 1995/96 near Framlingham, Suffolk Site 5 3 Nursery in England 1996/1997 Common Occasional 4 trees 8 2002/03 near Reepham, Norfolk 1 tree 9 2003/04 1 tree 10 2004/05 3 trees 11 2005/06 1 tree 12 2006/07 Site 6 8 Nursery in England 1996/1997 Common Occasional 4 trees 7 2001/02 near Reepham, Norfolk 5 trees 8 2002/03 1 tree 13 2007/08 1 tree 15 2009/10 1Categories are quantified as: Scarce < 1 per cent, Occasional 1–10 per cent, Common 11–50 per cent. 2After the main planting year some replanting was undertaken the following year to replace plants that failed to establish, so planting and year of death may be +1 year. View Large Table 2 Dating and PCR testing to detect the ash dieback pathogen Hymenoscyphus fraxineus applied to 27 diseased Fraxinus excelsior (ash) samples collected from three sites in England. Site number Sample code Tree height (m) Symptoms on main stem (& height above ground level) Planting year Age at planting Years of growth Year of tree death Years of growth around canker Year infection-causing canker initiated Outcome of RT-PCR-based detection of H. fraxineus 3 4 2 Five cankers, between 0.1–1.8 m 2001 21 13 2012/13 2 2010/11 Inconclusive 3 22 3.5 Stem 1: 6 cankers 40 cm–2 m. Stem 2: 5 cankers 2001 21 14 2013/14 3 2010/11 Inconclusive 3 23 2.5 Seven cankers, between 0.3–1.8 m 2001 21 13 2012/13 2 2010/11 Inconclusive 4 3 1.6 Multiple small cankers on stem and branches 1991 2 15 2004/05 0 2004/05 Positive 5 1 1.6 Large canker at 0.5 m and above. Epicormics at base 1996 2 14 2008/09 1 2007/08 Inconclusive 5 2 2.6 Five large cankers, between 0.2–0.8 m, plus basal lesions 1996 2 13 2007/08 1 2006/07 Negative 5 5 2.5 Two cankers, between 0.5–1.2 m 1996 2 14 2008/09 0 2008/09 Positive 5 6 2 Seven cankers, between 0.3–1.2 m 1996 2 12 2006/07 1 2005/06 Negative 5 7 2.2 Two coalesced cankers at 0.4 m 1996 2 14 2008/09 2 2006/07 Positive2 5 8 2.3 One small canker at 1.0 m 1996 2 13 2007/08 1(?) 2006/07 Negative 5 9 2.4 Eight cankers between 0.2–1.0 m 1996 2 17 2011/12 1 2010/11 Negative 5 10 3.5 Multiple cankers between 0.2–2.5 m 1996 2 17 2011/12 1(?) 2010/11 Positive 5 11 4.5 Five cankers, between 1.0–3.5 m 1996 2 15 2009/10 2 2007/08 Inconclusive 5 12 3 Numerous cankers between 1.0–2.7 m 1996 2 14 2008/09 1 2007/08 Negative 5 13 4.2 Four cankers between 1.5–3.0 m 1996 2 14 2008/09 2 2006/07 Negative 5 14 2.4 Numerous coalesced cankers between 0–1.2 m 1996 2 15 2009/10 1 2008/09 Positive2 5 15 5 Multiple cankers between 0.5–4.0 m 1996 2 17 2011/12 1 2010/11 Negative 5 16 2.5 Five cankers between 0.5–1.5 m 1996 2 14 2008/09 1 2007/08 Negative 5 17 2.5 Six cankers between 0.3–2.0 m 1996 2 14 2008/09 1 2007/08 Inconclusive 5 18 3.4 Numerous cankers between 0–2.0 m 1996 2 14 2008/09 1 2007/08 Inconclusive 5 19 4.7 Six cankers between 1.3–3.5 m 1996 2 14 2008/09 1 2007/08 Inconclusive 5 20 2.5 Three large cankers between 0.3–0.8 m 1996 2 14 2008/09 2 2006/07 Positive 5 21 3 Multiple cankers between 0–2.0 m 1996 2 14 2008/09 2 2006/07 Negative 5 24 3.2 Multiple cankers between 0.5–3.0 m 1996 2 15(14–18) 2009/10 (2008–13) 1 2008/09 Inconclusive 5 31 2 One canker at 0.7 m 1996 2 13 2007/08 2 2005/06 Negative 5 32 2.5 Three cankers above 1.2 m 1996 2 13 2007/08 1 2006/07 Negative 5 33 2 Three cankers between 0.5–1.5 m 1996 2 14 2008/09 1(?) 2007/08 Negative 5 34 Healthy stem material – – – Negative 5 35 Healthy stem material – – – Negative Site number Sample code Tree height (m) Symptoms on main stem (& height above ground level) Planting year Age at planting Years of growth Year of tree death Years of growth around canker Year infection-causing canker initiated Outcome of RT-PCR-based detection of H. fraxineus 3 4 2 Five cankers, between 0.1–1.8 m 2001 21 13 2012/13 2 2010/11 Inconclusive 3 22 3.5 Stem 1: 6 cankers 40 cm–2 m. Stem 2: 5 cankers 2001 21 14 2013/14 3 2010/11 Inconclusive 3 23 2.5 Seven cankers, between 0.3–1.8 m 2001 21 13 2012/13 2 2010/11 Inconclusive 4 3 1.6 Multiple small cankers on stem and branches 1991 2 15 2004/05 0 2004/05 Positive 5 1 1.6 Large canker at 0.5 m and above. Epicormics at base 1996 2 14 2008/09 1 2007/08 Inconclusive 5 2 2.6 Five large cankers, between 0.2–0.8 m, plus basal lesions 1996 2 13 2007/08 1 2006/07 Negative 5 5 2.5 Two cankers, between 0.5–1.2 m 1996 2 14 2008/09 0 2008/09 Positive 5 6 2 Seven cankers, between 0.3–1.2 m 1996 2 12 2006/07 1 2005/06 Negative 5 7 2.2 Two coalesced cankers at 0.4 m 1996 2 14 2008/09 2 2006/07 Positive2 5 8 2.3 One small canker at 1.0 m 1996 2 13 2007/08 1(?) 2006/07 Negative 5 9 2.4 Eight cankers between 0.2–1.0 m 1996 2 17 2011/12 1 2010/11 Negative 5 10 3.5 Multiple cankers between 0.2–2.5 m 1996 2 17 2011/12 1(?) 2010/11 Positive 5 11 4.5 Five cankers, between 1.0–3.5 m 1996 2 15 2009/10 2 2007/08 Inconclusive 5 12 3 Numerous cankers between 1.0–2.7 m 1996 2 14 2008/09 1 2007/08 Negative 5 13 4.2 Four cankers between 1.5–3.0 m 1996 2 14 2008/09 2 2006/07 Negative 5 14 2.4 Numerous coalesced cankers between 0–1.2 m 1996 2 15 2009/10 1 2008/09 Positive2 5 15 5 Multiple cankers between 0.5–4.0 m 1996 2 17 2011/12 1 2010/11 Negative 5 16 2.5 Five cankers between 0.5–1.5 m 1996 2 14 2008/09 1 2007/08 Negative 5 17 2.5 Six cankers between 0.3–2.0 m 1996 2 14 2008/09 1 2007/08 Inconclusive 5 18 3.4 Numerous cankers between 0–2.0 m 1996 2 14 2008/09 1 2007/08 Inconclusive 5 19 4.7 Six cankers between 1.3–3.5 m 1996 2 14 2008/09 1 2007/08 Inconclusive 5 20 2.5 Three large cankers between 0.3–0.8 m 1996 2 14 2008/09 2 2006/07 Positive 5 21 3 Multiple cankers between 0–2.0 m 1996 2 14 2008/09 2 2006/07 Negative 5 24 3.2 Multiple cankers between 0.5–3.0 m 1996 2 15(14–18) 2009/10 (2008–13) 1 2008/09 Inconclusive 5 31 2 One canker at 0.7 m 1996 2 13 2007/08 2 2005/06 Negative 5 32 2.5 Three cankers above 1.2 m 1996 2 13 2007/08 1 2006/07 Negative 5 33 2 Three cankers between 0.5–1.5 m 1996 2 14 2008/09 1(?) 2007/08 Negative 5 34 Healthy stem material – – – Negative 5 35 Healthy stem material – – – Negative 1Planting year estimated. 2Presence of Hymenoscyphus fraxineus confirmed for these samples by both RT-PCR and end-point PCR and ITS sequence analysis. View Large Table 2 Dating and PCR testing to detect the ash dieback pathogen Hymenoscyphus fraxineus applied to 27 diseased Fraxinus excelsior (ash) samples collected from three sites in England. Site number Sample code Tree height (m) Symptoms on main stem (& height above ground level) Planting year Age at planting Years of growth Year of tree death Years of growth around canker Year infection-causing canker initiated Outcome of RT-PCR-based detection of H. fraxineus 3 4 2 Five cankers, between 0.1–1.8 m 2001 21 13 2012/13 2 2010/11 Inconclusive 3 22 3.5 Stem 1: 6 cankers 40 cm–2 m. Stem 2: 5 cankers 2001 21 14 2013/14 3 2010/11 Inconclusive 3 23 2.5 Seven cankers, between 0.3–1.8 m 2001 21 13 2012/13 2 2010/11 Inconclusive 4 3 1.6 Multiple small cankers on stem and branches 1991 2 15 2004/05 0 2004/05 Positive 5 1 1.6 Large canker at 0.5 m and above. Epicormics at base 1996 2 14 2008/09 1 2007/08 Inconclusive 5 2 2.6 Five large cankers, between 0.2–0.8 m, plus basal lesions 1996 2 13 2007/08 1 2006/07 Negative 5 5 2.5 Two cankers, between 0.5–1.2 m 1996 2 14 2008/09 0 2008/09 Positive 5 6 2 Seven cankers, between 0.3–1.2 m 1996 2 12 2006/07 1 2005/06 Negative 5 7 2.2 Two coalesced cankers at 0.4 m 1996 2 14 2008/09 2 2006/07 Positive2 5 8 2.3 One small canker at 1.0 m 1996 2 13 2007/08 1(?) 2006/07 Negative 5 9 2.4 Eight cankers between 0.2–1.0 m 1996 2 17 2011/12 1 2010/11 Negative 5 10 3.5 Multiple cankers between 0.2–2.5 m 1996 2 17 2011/12 1(?) 2010/11 Positive 5 11 4.5 Five cankers, between 1.0–3.5 m 1996 2 15 2009/10 2 2007/08 Inconclusive 5 12 3 Numerous cankers between 1.0–2.7 m 1996 2 14 2008/09 1 2007/08 Negative 5 13 4.2 Four cankers between 1.5–3.0 m 1996 2 14 2008/09 2 2006/07 Negative 5 14 2.4 Numerous coalesced cankers between 0–1.2 m 1996 2 15 2009/10 1 2008/09 Positive2 5 15 5 Multiple cankers between 0.5–4.0 m 1996 2 17 2011/12 1 2010/11 Negative 5 16 2.5 Five cankers between 0.5–1.5 m 1996 2 14 2008/09 1 2007/08 Negative 5 17 2.5 Six cankers between 0.3–2.0 m 1996 2 14 2008/09 1 2007/08 Inconclusive 5 18 3.4 Numerous cankers between 0–2.0 m 1996 2 14 2008/09 1 2007/08 Inconclusive 5 19 4.7 Six cankers between 1.3–3.5 m 1996 2 14 2008/09 1 2007/08 Inconclusive 5 20 2.5 Three large cankers between 0.3–0.8 m 1996 2 14 2008/09 2 2006/07 Positive 5 21 3 Multiple cankers between 0–2.0 m 1996 2 14 2008/09 2 2006/07 Negative 5 24 3.2 Multiple cankers between 0.5–3.0 m 1996 2 15(14–18) 2009/10 (2008–13) 1 2008/09 Inconclusive 5 31 2 One canker at 0.7 m 1996 2 13 2007/08 2 2005/06 Negative 5 32 2.5 Three cankers above 1.2 m 1996 2 13 2007/08 1 2006/07 Negative 5 33 2 Three cankers between 0.5–1.5 m 1996 2 14 2008/09 1(?) 2007/08 Negative 5 34 Healthy stem material – – – Negative 5 35 Healthy stem material – – – Negative Site number Sample code Tree height (m) Symptoms on main stem (& height above ground level) Planting year Age at planting Years of growth Year of tree death Years of growth around canker Year infection-causing canker initiated Outcome of RT-PCR-based detection of H. fraxineus 3 4 2 Five cankers, between 0.1–1.8 m 2001 21 13 2012/13 2 2010/11 Inconclusive 3 22 3.5 Stem 1: 6 cankers 40 cm–2 m. Stem 2: 5 cankers 2001 21 14 2013/14 3 2010/11 Inconclusive 3 23 2.5 Seven cankers, between 0.3–1.8 m 2001 21 13 2012/13 2 2010/11 Inconclusive 4 3 1.6 Multiple small cankers on stem and branches 1991 2 15 2004/05 0 2004/05 Positive 5 1 1.6 Large canker at 0.5 m and above. Epicormics at base 1996 2 14 2008/09 1 2007/08 Inconclusive 5 2 2.6 Five large cankers, between 0.2–0.8 m, plus basal lesions 1996 2 13 2007/08 1 2006/07 Negative 5 5 2.5 Two cankers, between 0.5–1.2 m 1996 2 14 2008/09 0 2008/09 Positive 5 6 2 Seven cankers, between 0.3–1.2 m 1996 2 12 2006/07 1 2005/06 Negative 5 7 2.2 Two coalesced cankers at 0.4 m 1996 2 14 2008/09 2 2006/07 Positive2 5 8 2.3 One small canker at 1.0 m 1996 2 13 2007/08 1(?) 2006/07 Negative 5 9 2.4 Eight cankers between 0.2–1.0 m 1996 2 17 2011/12 1 2010/11 Negative 5 10 3.5 Multiple cankers between 0.2–2.5 m 1996 2 17 2011/12 1(?) 2010/11 Positive 5 11 4.5 Five cankers, between 1.0–3.5 m 1996 2 15 2009/10 2 2007/08 Inconclusive 5 12 3 Numerous cankers between 1.0–2.7 m 1996 2 14 2008/09 1 2007/08 Negative 5 13 4.2 Four cankers between 1.5–3.0 m 1996 2 14 2008/09 2 2006/07 Negative 5 14 2.4 Numerous coalesced cankers between 0–1.2 m 1996 2 15 2009/10 1 2008/09 Positive2 5 15 5 Multiple cankers between 0.5–4.0 m 1996 2 17 2011/12 1 2010/11 Negative 5 16 2.5 Five cankers between 0.5–1.5 m 1996 2 14 2008/09 1 2007/08 Negative 5 17 2.5 Six cankers between 0.3–2.0 m 1996 2 14 2008/09 1 2007/08 Inconclusive 5 18 3.4 Numerous cankers between 0–2.0 m 1996 2 14 2008/09 1 2007/08 Inconclusive 5 19 4.7 Six cankers between 1.3–3.5 m 1996 2 14 2008/09 1 2007/08 Inconclusive 5 20 2.5 Three large cankers between 0.3–0.8 m 1996 2 14 2008/09 2 2006/07 Positive 5 21 3 Multiple cankers between 0–2.0 m 1996 2 14 2008/09 2 2006/07 Negative 5 24 3.2 Multiple cankers between 0.5–3.0 m 1996 2 15(14–18) 2009/10 (2008–13) 1 2008/09 Inconclusive 5 31 2 One canker at 0.7 m 1996 2 13 2007/08 2 2005/06 Negative 5 32 2.5 Three cankers above 1.2 m 1996 2 13 2007/08 1 2006/07 Negative 5 33 2 Three cankers between 0.5–1.5 m 1996 2 14 2008/09 1(?) 2007/08 Negative 5 34 Healthy stem material – – – Negative 5 35 Healthy stem material – – – Negative 1Planting year estimated. 2Presence of Hymenoscyphus fraxineus confirmed for these samples by both RT-PCR and end-point PCR and ITS sequence analysis. View Large To gather evidence of how long symptoms of ash dieback might have been visible on each site, all six sites were surveyed in detail in 2013. For each site outside East Anglia, a wider 1.5 km radius survey was also undertaken, to ascertain extent and distribution of symptoms in the wider environment. Efforts were made to look for factors that were indicators of longstanding disease including widespread dieback, tree mortality, presence of lesions at the tree collar often referred to as basal lesions (Husson et al., 2012) and whether H. fraxineus apothecia were abundant in the litter layer under symptomatic trees. When dead trees were located, evidence of H. fraxineus in the form of diamond-shaped cankers on the main stem was recorded (i.e. those likely to have been killed by ash dieback). Any dead trees with visible cankers were cut close to the base to reveal the stem cross-section and a count was made of the number of annual rings by eye, to establish an estimated year of death using the known planting date for each site. Planting year and origin of tree stock Evidence of the year of planting and the trees source was sought for each of the selected sites. Site owners, where available, were asked for relevant information, and Forestry Commission England’s database was interrogated for further information on grant assistance for each site. Additionally, aerial photographs of each site from spring/summer 1999 were visually analysed for verification. Where very limited information was available (site 3), a live tree from the same compartment was felled and the ring-count information used to determine likely planting year (see below). Tree dating At three of the sites a more rigorous ring-count analysis was undertaken (Table 2); samples were taken from 27 dead trees, consisting of basal discs and lengths of the main stem, each with one or more H. fraxineus-like cankers. All discs or stem sections were sealed individually in a plastic bag, labelled and returned to the laboratory for analysis. On receipt the samples were recorded and relabelled with deep-stapled metal tags; basal discs were then oven dried immediately, while stem sections were stored at 4°C pending molecular analyses as described in the next section. For accurate counts of the annual growth rings in the 27 trees, basal discs were dried at 50°C over several days until they reached a constant weight, and then one surface on each disc was sanded to create a smooth, level surface so the growth rings could be counted. As the planting date of each tree was known, date of tree death could be calculated based on the ring counts. Annual growth rings were assessed only by presence; no quantification of increment size was attempted although reductions were usually apparent in the final 1–2 years of the tree’s life Once the stem sections with ash dieback cankers had been sampled for use in PCR detection of H. fraxineus (see below), each section was cross-cut at the widest part of the canker, then dried and sanded in the same way as the basal discs. Each section was also photographed (Figure 2), and the callus growth around the edge of the canker was examined to establish the year in which phloem and cambium was first damaged and therefore how many years prior to tree death the canker symptoms first appeared. Figure 2 View largeDownload slide Canker symptoms likely to be caused by Hymenoscyphus fraxineus on dead ash trees planted in 1996 and detailed in Table 2. Cankers on trees that died in: (a) 2012/2013 (sample 4), (b) 2011/2012 (sample 10), (c) 2009/2010 (sample 11), (d) 2008/2009 (sample 13), (e) 2009/2010 (sample 14), (f) 2011/2012 (sample 15), (g) 2008/2009 (sample 17), (h) 2008/2009 (sample 20), (i) 2008/2009 (sample 21), (j) 2007/2008 (sample 32). Figure 2 View largeDownload slide Canker symptoms likely to be caused by Hymenoscyphus fraxineus on dead ash trees planted in 1996 and detailed in Table 2. Cankers on trees that died in: (a) 2012/2013 (sample 4), (b) 2011/2012 (sample 10), (c) 2009/2010 (sample 11), (d) 2008/2009 (sample 13), (e) 2009/2010 (sample 14), (f) 2011/2012 (sample 15), (g) 2008/2009 (sample 17), (h) 2008/2009 (sample 20), (i) 2008/2009 (sample 21), (j) 2007/2008 (sample 32). Additional sampling and analysis A single tree was selected a few metres away from the pocket of ash with extensive and severe symptoms of ash dieback at site 3. It had several well developed cankers at the base of the main stem which extended up to 1.5 m from the base. The extent of callusing around the canker margins suggested that the cankers had been initiated some years earlier. The tree was felled in January 2015 and approximately 1 m of the cankered stem returned to the laboratory for analysis and dating. Prior to dating of the cankers, isolation of H. fraxineus was attempted from the necrotic phloem tissue of each discrete stem lesions using methods of EPPO (2013). Isolates obtained were confirmed as H. fraxineus based on colony morphology and the PCR diagnostic of Johansson et al. (2010). Detection of Hymenoscyphus fraxineus in cankers All stem canker samples from dead trees were bagged and stored separately at 4°C to prevent any cross-contamination. In addition, two ‘healthy’ stem sections (i.e. with no external lesions) were collected from site 5 to act as controls. An isolated room with no history of H. fraxineus diagnosis was used to process the samples. Prior to this, and to avoid any possibility of contamination with H. fraxineus fungal material or DNA, the room was thoroughly disinfected with both 70 per cent ethanol (v/v) and 10 per cent sodium hypochlorite solution (v/v; bleach). All the stem sections were also rinsed with 70 per cent ethanol for 1 min, followed by bleach solution for 1 min, and finished by a wash in sterile distilled water for 1 min, after which they were air-dried at room temperature. A mallet and chisel (wiped with bleach and flame dried) were used to split logs longitudinally, and a sliver of underlying xylem tissue beneath the canker was excised using a similarly cleaned scalpel. Each xylem sample was then transferred to a sterile tube and stored at −20°C, until subsequent DNA extraction using a DNAmite kit (Microzone Ltd, UK) into a volume of 300 μl. For subsequent use in real-time PCR (RT-PCR) testing, 100 μl of the DNA extract was purified using a OneStepTM PCR Inhibitor Removal Kit (Zymo Research, USA). Real-time PCR analysis All 29 DNA samples (27 from symptomatic material, 2 healthy controls) were tested for the presence of H. fraxineus using species-specific multiplex RT-PCR (EPPO, 2013; Table 2), with two technical replicates of each DNA extract tested. RT-PCR simultaneously targeted both H. fraxineus and host plant DNA using different primer/probe combinations (Ioos et al., 2009; Ioos and Fourrier, 2011) and used a Roche Lightcycler 480 instrument (Roche Diagnostics, Basel, Switzerland). Data analysis used the LightCycler 480 (version 1.5.0.39) software with the absolute quantification second derivatives maximum method used to calculate the crossing point (Cp) values. Samples that produced Cp values of <40 for both technical replicates targeted to H. fraxineus DNA were considered positive for H. fraxineus. Samples for which only a single replicate produced a Cp value of <40 were considered inconclusive. Samples that produced Cp values of <40 for both technical replicates targeted to host plant DNA (but with neither replicate producing a signal for H. fraxineus) were considered negative. Both positive (H. fraxineus and ‘healthy’ plant template DNA of known Cp value) and negative (no-template) water controls were included in all testing. Results Site observations Site 1: Near Wooler, Northumberland This isolated c. 1.2 ha stand of almost pure ash was planted between 1999 and 2001. Owner records provided planting year and details of the supplying nursery; the latter confirmed these records and the age of stock supplied (Tables 1 and 2). The supplying nursery was in England but was known to occasionally supplement stock by purchasing trees from a wholesaler which routinely imported tree stock from mainland Europe. Planting positions were clearly visible in 1999 aerial photographs. Initial investigation revealed one area of the stand with notably poorer growth. There was little mortality and the two trees that died in ca. 2001 (Table 1) were so decayed that cause of death was inconclusive. Ash dieback symptoms were most developed in the area of poorest growth and some naturally regenerating ash trees in this area also had limited symptoms, indicating recent infection. All trees were established in tree shelters so no assessment for basal lesions was made. Apothecia were found readily under the symptomatic trees with poorer growth and less frequently elsewhere. Occasional shoot dieback was observed on some mature trees immediately adjacent to the planted stand suggesting recent disease spread. The 1.5 km radius survey around the site revealed sporadic symptoms in mature, young and naturally regenerating ash trees in contiguous woodland to the north of the site up to 1.2 km away. No symptoms were identified in the many hedgerow trees dividing fields to the east of the site, and no infection was detected on younger recently planted ash stands 800 m to the north. Site 2: Near Bickleigh, Devon The 28 ha site comprised pure blocks of various broadleaf species originally planted in 1996, with additional planting in 2008; planting year and plant origins were provided by the owner (Tables 1 and 2). Planting positions were clearly visible in 1999 aerial photographs. Initial inspection suggested that a high percentage of the more recently (2008)-planted ash had symptoms consistent with H. fraxineus infection including stem/branch cankers and wilting, although apothecia were not easily found on fallen rachises under these trees (A. Whybrow and B. Jones, pers. comm). In contrast, the ash trees planted in 1996/1997, and also a single block that had been replanted to replace failed cherry trees, showed heavy dieback with many apothecia under these trees. Basal lesions were also common on trees in the same block but seen only infrequently in adjoining areas of ash. Overall, mortality was low and estimated year of death was between 2004 and 2010 (Table 1). The 1.5 km survey revealed sporadic symptoms in hedgerows and in an area of recently planted ash almost 1.5 km away, with the highest proportion of symptoms in trees to the north-west of the site. Symptoms were also occasionally identified in neighbouring mature ash woodlands. Another isolated site (but with more recent signs of infection) was found over 4.5 km away to the south east. Site 3: Near Swadlincote, Derbyshire/Leicestershire border This 27 ha broadleaf plantation (planted from 1996 to 2008) contained older stands of ash planted in small groups, and younger plantings of ash in blocks of varying size. Symptoms on this site had been reported to be worst on trees planted in 2008. However, closer inspection revealed that although some trees planted several years earlier had only recently become infected, there were other pockets of similarly aged ash with severe symptoms, including a small number of trees with basal lesions suggesting a longstanding disease presence. The most extensive dieback was concentrated in a group of approximately 20 ash trees, planted within a compartment of mixed broadleaved trees and adjacent to heavily affected areas of 2008-planted ash which were downhill. Mortality in this pocket was confined to seven trees which appeared to have died between 2007 and 2012 (Table 1). A further three trees sampled during a later visit were found to have died between 2012 and 2014 (Table 2). Sporadic ash planted over a similar timescale were found with H. fraxineus infection almost to the edge of the 1.5 km survey radius. Most of the symptomatic trees were located to the east of the site, following the direction of prevailing winds. Site 4: Near Framlingham, Suffolk This extensive newly established 90 ha broadleaf wood was created between 1991 and 2002 on farmland which surrounded small pockets of long established ash woodland. The owner was able to confirm the planting years of compartments from site records. The supplying nursery was no longer in business, but anecdotal information suggested that it occasionally purchased stock from various sources including from mainland Europe. Established trees were clearly visible in 1999 aerial photographs; the ash had generally been planted in small pure blocks. The amount of dieback across the site varied, but basal lesions were only occasionally observed. Although investigations initially concentrated on areas identified by Forestry Commission England aerial surveillance as those with the greatest levels of dieback, there was no single area where symptoms were more developed and therefore where the disease might have been present longer than elsewhere. Only a single dead tree with typical ash dieback symptoms was identified (Table 1). Apothecia were found readily on rachises across the large site, including under the mature canopy of one small ancient ash stand where they sometimes occurred on small woody material in the leaf litter rather than on rachises. A formal 1.5 km survey was not undertaken as this site was one of the many identified in East Anglia over the autumn/winter 2012 Forestry Commission surveys. However, other nearby sites were visited and confirmed to be infected as part of this in depth investigation. One dead tree was identified that had died between 2004 and 2005 (Table 2) Site 5: Near Reepham, Norfolk This small 3 ha site was planted in 1996, and was originally owned, established from the same stock and managed by the same estate as site 6. The two sites are approximately 400 m apart, separated by a disused railway, open farmland and fragments of woodland, including older ash stands and younger, more recently planted ash. The owner was able to supply details of planting year and the supplying nursery. The latter also confirmed the age of stock supplied (Tables 1 and 2) and confirmed that any Fraxinus stock supplied would have been grown on the nursery from seed. Planting positions were clearly visible in the 1999 aerial photographs. Aerial surveillance of sites 5 and 6 revealed worse levels of dieback than elsewhere in the area, and the extent of dieback suggested a high level of mortality. However, ground inspection revealed most trees were still alive. Dead trees were scattered across the site, and some larger individuals appeared to have been entirely girdled by H. fraxineus basal lesions. Apothecia and sclerotized rachises were also abundant across the site. Most of the dead trees were between 1 and 3 m in height, and initially 10 of the smallest dead trees with obvious cankers were dated as having died between 2002 and 2007 (Table 1). A further 23 trees sampled during a later visit were found to have died between 2006 and 2012 (Table 2). As with site 4, a formal 1.5 km survey was never undertaken. However, inspections of adjacent ash stands revealed an absence of similarly well established symptoms, while nearby mature and ancient stands of ash did not have high levels of dieback and only a few apothecia were found underneath them. Site 6: Near Reepham, Norfolk This 8 ha site that was established at the same time as site 5 in 1996, and had similar site conditions. Details of the planting date and stock origins were the same as site 5. Subsequently the site had been sold and had not been thinned. There was limited mortality and most of the dead trees were 1–3 m in height and estimated year of death ranged from 2001 to 2010 (Table 1). Many near-dead, smaller trees had been girdled but still had epicormic shoots close to their root collars but largely hidden by the grass sward. Eleven of the smallest dead trees with typical ash dieback cankers were sampled (Table 1). Many others had basal cankers consistent with longstanding H. fraxineus infection, and when some were felled for investigation, it was clear that incremental growth had been reducing over a number of years after establishment, suggestive of long-term infection. Another indicator of a long established outbreak was abundant apothecia on both fallen rachises and small sections of woody material on the forest floor (Kirisits et al., 2014). Tree dating Table 1 shows the data for the 36 dead trees that were initially identified for investigation at the six sites. Most of the planting stock had been obtained from UK nurseries and only the stock for the Devon site (site 2) had definitely come directly from a non-UK source. Nursery and/or estate records for sites 1, 2, 4, 5 and 6 confirmed that all plants were two years old when planted out and probably stock for site 3 would have been the same. Replacement stock was planted the year following establishment to make good any planting failures. Apart from site 3, all the replacement plants were known to be sourced from the same nurseries that supplied the original stock. With the exception of site 1, all the sites had at least one tree that had died some years earlier with one or more diamond-shaped canker(s) on the stem consistent with the symptoms caused by H. fraxineus. Although the planting date of most of the cankered dead trees was likely to be the initial establishment year, it could not be ruled out that some of the trees might be one year younger (i.e. were replacement stock). However, based on the ring-count assessments made in the field and the known planting dates for each site, the earliest year of tree death was estimated at 1995/96 (site 4) and the most recent in 2011/12 (site 3), with 19 out of the total of 36 sampled trees estimated to have died by 2004/05. When a more detailed tree dating analysis was undertaken using a further 27 dead trees taken from sites 3, 4 and 5 (Table 2) a similar pattern emerged, although the accurate counts of growth rings made in the laboratory indicated that the ring counts made in the field had underestimated the age of the trees and therefore the year in which they died. As before, each of the 27 dead trees had at least one diamond-shaped canker on the stem indicative of H. fraxineus infection. Examples of the cankers are shown in Figure 2. With this second set of trees, ring counts revealed that 18 out of the 27 had died by 2008/2009, which corresponded to between 10 and 13 years after being planted out. Dating the callus growth on the trees revealed cankers had been initiated up to 3 years earlier, so the symptoms of ash dieback symptoms could be dated at least as far back as 2004/2005 on one tree and to 2005/2006 for a further two (Table 2). Additional sampling and analysis Dating the ~1.0 m length of stem taken from a tree at site 3 in 2015 (close to the area of greatest symptom intensity and tree mortality), indicated the first growth occurred in 2000. Thus assuming a planting stock age of 2 years, this suggests the likely planting date for the compartment at site 3 of 2001/2. The same section of stem had six sunken cankers, two of which had coalesced and each canker ranged in length from ca. 12 to 25 cm. Hymenoscyphus fraxineus was isolated from four of the six cankers demonstrating they were active lesions, and it transpired that all H. fraxineus isolates were of the same mating type (MAT1-1). Despite the sunken and aged appearance of the cankers which were surrounded by high ridges of callus tissue, this proved to be deceptive when estimating canker age. The depth of the cankers was largely due to vigorous growth and formation of wide growth rings between 2012 and 2014 from the uninjured cambium and phloem rather than multiple smaller growth rings over a longer period of time. Overall, four of the active lesions dated from 2012; one lesion for which H. fraxineus could not be isolated (inactive lesion) dated from 2010; and another inactive lesion dated from 2011 at the latest (Table 3). This suggests that the fungus is capable of remaining alive and colonizing stem material for upwards of 3–4 years after growing into bark tissue from infected shoots or foliage. Table 3 Timing of lesion development by Hymenoscyphus fraxineus on a single live Fraxinus excelsior log collected in 2015. Lesion number Date of lesion initiation Earliest year of foliar infection H. fraxineus isolated Lesion 1 2012 2011 Yes Lesion 2A 2012 2011 Yes Lesion 2B 2012 2011 Yes Lesion 3 2012 2011 Yes Lesion 4 2010 2009 No Lesion 5 2011 (at latest1) 2010 No Lesion number Date of lesion initiation Earliest year of foliar infection H. fraxineus isolated Lesion 1 2012 2011 Yes Lesion 2A 2012 2011 Yes Lesion 2B 2012 2011 Yes Lesion 3 2012 2011 Yes Lesion 4 2010 2009 No Lesion 5 2011 (at latest1) 2010 No 1Log not cross-sectioned at lesion centre, so infection could have been earlier. Table 3 Timing of lesion development by Hymenoscyphus fraxineus on a single live Fraxinus excelsior log collected in 2015. Lesion number Date of lesion initiation Earliest year of foliar infection H. fraxineus isolated Lesion 1 2012 2011 Yes Lesion 2A 2012 2011 Yes Lesion 2B 2012 2011 Yes Lesion 3 2012 2011 Yes Lesion 4 2010 2009 No Lesion 5 2011 (at latest1) 2010 No Lesion number Date of lesion initiation Earliest year of foliar infection H. fraxineus isolated Lesion 1 2012 2011 Yes Lesion 2A 2012 2011 Yes Lesion 2B 2012 2011 Yes Lesion 3 2012 2011 Yes Lesion 4 2010 2009 No Lesion 5 2011 (at latest1) 2010 No 1Log not cross-sectioned at lesion centre, so infection could have been earlier. Real-time PCR and end-point PCR Of the 27 canker samples tested with RT-PCR, H. fraxineus was confirmed in six (samples 3, 5, 7, 10, 14 and 20) and the two visually healthy shoots were negative for H. fraxineus (Table 2). Of the remaining samples, twelve were negative for H. fraxineus and the remaining nine produced an inconclusive result (Table 2). Additional definitive evidence for H. fraxineus was obtained in two more samples (7 and 14) with less sensitive conventional PCR using the primers of Johansson et al. (2010); partial ITS sequences obtained from these samples both showed 100 per cent identity to the type of H. fraxineus (Oth_01; GenBank Acc: GU586904). Discussion To evaluate whether ash planted over the previous 20 years might have included any plants infected by the ash dieback pathogen, H. fraxineus, certain criteria were required when selecting sites for study. Ideally, sites needed to be relatively isolated and therefore unlikely to be affected by airborne H. fraxineus inoculum originating from any nearby local disease foci initiated either from very recently planted infected ash or from disease outbreaks on mainland Europe. If possible, study sites were also excluded with multiple planting episodes over the life of the plantation, to reduce the opportunities for the oldest trees on the site to be cross-infected by diseased stock introduced several years later. Although the six sites selected for study (established between 1991 and 2002) did not meet all these criteria completely, we were able to find evidence of infection by H. fraxineus that dated back at least to the mid-2000s. Moreover, based on the site characteristics, the most parsimonious explanation for the established infection is that it originated from use of infected planting stock rather than from airborne inoculum. In particular, the location and geographical isolation of sites 1, 2 and 3 in Northumberland, Leicestershire and Devon, made the likelihood of exposure to airborne ascospores from ash dieback outbreaks in mainland Europe very small. Undertaking accurate counts of annual growth rings in the laboratory to determine the year of tree death as well as the year in which cankers were initiated on the main stem of affected trees, allowed disease symptoms to be dated back to at least 2004/2005 on a site planted in 1991/1992 and to 2005/2006 on a site planted in 1996. Ring dating undertaken in the field also indicated that the disease was longstanding but gave less accurate results for the date of first canker symptoms, possibly because of the difficulty in distinguishing individual annual rings from early and late stage annual growth. In addition, with older samples the dead wood tended to be of poor quality again affecting count accuracy. Thomsen and Jorgensen (2011) reported that the size of growth of rings can reduce markedly following high infection levels of H. fraxineus again making such assessments difficult. Molecular analyses also confirmed the presence of H. fraxineus in about 22 per cent of the symptomatic trees, at least one of which had died in 2004/2005 and three more which had died in 2008/2009 and had canker symptoms that had been initiated one or two years earlier. The failure to detect H. fraxineus in more of the samples with the typical ash dieback cankers that died prior to 2008 could have several explanations. It may be that H. fraxineus was not present in those samples and the symptoms were caused by other canker causing agents that affect ash (Kowalski and Lukomska, 2005). There is also the prospect that the DNA extracted from older symptomatic material had become too degraded to amplify and detect following invasion by wood decay fungi and bacteria, or over time the wood had accumulated inhibitors which affected the PCR process. Either is a possibility given the failure in RT-PCR testing to detect both plant DNA as well as H. fraxineus DNA from some of the sampled plant material. Despite this, the combined evidence of typical symptoms of ash dieback that could be reliably dated back to the mid-2000s and PCR-based confirmation of the presence of H. fraxineus in some of this material, indicate that the visible signs of ash dieback had started to occur almost a decade before the first formal reports of ash dieback in the UK in 2012. If ash dieback was indeed present on some ash stock planted as early as 1991, it raises the question of why disease development appears to have been relatively slow. Only low levels of mortality were observed some 12–18 years after planting and disease spread to neighbouring woodland and hedgerow ash trees was limited based on the findings of the 1.5 km radius surveys. Assuming that H. fraxineus was introduced to each site on infected planting stock and laboratory analysis dates some ash dieback cankers to 2004/2005, it follows that there must have been pseudosclerotial material for the formation of apothecia and some spore production on site at least in the previous year (2003/2004) and probably earlier. However, earlier cankers are likely to have been present on smaller dead plants but easily overlooked. In addition, by time of investigation such material would be too degraded for analysis (e.g. site 1, Table 1), or restricted to fine shoots which would have broken off and disintegrated long ago. Studies in nurseries, however, have shown that pseudosclerotia giving rise to apothecia can form at the root collars and lower stems of infected ash plants after death (Kowalski and Holdenrieder, 2009a,b; Kirisits et al., 2012), suggesting that some infected saplings with H. fraxineus lesions could sustain spore production leading to disease spread. A similar observation was made in Carmarthenshire, where apothecia were found on a basal lesion caused by H. fraxineus (Figure 3). Field observations in England have also shown that apothecia can be found on woody shoot material in the litter layer (M Biddle, B Wylder, J. Webber, unpublished), and Kirisits et al. (2014) showed that a small proportion of ash shoots with fresh H. fraxineus lesions which were then exposed naturally by placing on the soil surface for a year could go on to form apothecia. Figure 3 View largeDownload slide Fraxinus excelsior tree planted in 2007 at a site in Carmarthenshire found with apothecia growing from a Hymenoscyphus fraxineus basal lesion in 2014. Image supplied courtesy of Menna Langford, Natural Resources Wales. Figure 3 View largeDownload slide Fraxinus excelsior tree planted in 2007 at a site in Carmarthenshire found with apothecia growing from a Hymenoscyphus fraxineus basal lesion in 2014. Image supplied courtesy of Menna Langford, Natural Resources Wales. The time between planting of infected stock and a spreading ash dieback outbreak will also be dependent on the amount of time taken for apothecial formation. This requires both mating types of the fungus to be present and come into physical contact with each other so that sexual reproduction can occur (Gross et al., 2012) and only then can apothecial formation occur sometime later. However, even when compatible mating types are present, initial levels of ascospore production are likely be very low. The formation of new H. fraxineus cankers on the main stems of affected trees may also be delayed as much of the infected foliage is now known to be shed before the pathogen can extend into the bark of shoots and stems to cause girdling cankers (Kirisits and Freinschlag, 2012; Krautler and Kirisits, 2012). In addition it has been recognized that incidence and severity of dieback observed at different locations in mainland Europe can vary depending on site-specific ecological conditions (Pliūra et al., 2011; Schumacher, 2011; Skovsgaard et al., 2017). To a great extent this probably relates to how suitable climatic conditions are for apothecia production and ascospore release (Gross and Holdenreider, 2013). The characteristics of individual sites may also influence disease spread. For example, many of the newly established ash sites in England are on farmland where the ground cover is initially bare soil that later develops into a vigorous grass sward which envelops any pseudosclerotial rachises and may hinder apothecia formation and spore release. It can also take more than one year before infected rachises and shoot material go on to produce apothecia of H. fraxineus, especially if environmental conditions are unfavourable (Gross and Holdenreider, 2013; Kirisits et al., 2014). The evidence that H. fraxineus was present and infecting ash in England as early as 2004/2005 not only pre-dates by a number of years the first formal record of the pathogen in the UK but also the earliest acknowledgements of a new pathogen in other European countries (Sansford, 2013). It also pre-dates much of our understanding of the pathogen life cycle of the fungus (summarized by Gross et al., 2014), and the naming of the causal agent of ash dieback by Kowalski (2006). The significance of lesions or other symptoms on planting stock in the UK are therefore unlikely to have been recognized or reported. It is also notable, that out of the six sites that we studied, five were apparently stocked with ash plants from British nurseries and not from directly imported plants from mainland Europe where the disease is acknowledged to have been established for much longer. The practice of sourcing ash seed in the UK and then exporting it to mainland Europe for plant production before importing the saplings back to the UK for growing on and planting (Downing, 2012) could explain this discrepancy. However, even if a ban on the movement of ash from mainland Europe into the UK had been put in place immediately after the first description of the new pathogen in mainland Europe (Kowalski, 2006) rather than in October 2012 after the first findings in the wider environment in England, our study indicates that infected ash plants were already in the UK. Once sporulation of H. fraxineus began to take place at geographically diverse locations across the UK, the spread of the fungus was likely to be inevitable. Conclusions Our evidence indicates that ash dieback has been present in the UK much earlier than previously understood – prior even to the first description of the pathogen in mainland Europe in 2006. Our observations suggest there may be a lag period of several years from the time of introduction on infected plants to an obvious wider environment outbreak of ash dieback. Sporulation from these infected plants must have occurred to allow disease spread, probably through the formation of apothecia on the woody stems of affected plants. This process is likely to have occurred many times at other sites where infected ash has been planted in the UK. Hymenoscyphus fraxineus can remain active in stem lesions 3–4 years after infection despite an active callus response from infected trees. It proved possible to extract and detect DNA of H. fraxineus from old cankers, in one instance up to twelve years after tree death, but more often material of that age (and younger) was too degraded for successful detection of the pathogen. Acknowledgements We thank colleagues at Forest Research including Prof. Clive Brasier for advice on site investigation and Andrew Jeeves for careful preparation of ash samples for dating. We also thank our colleagues from Forestry Commission England, Alan Ockenden, Emily Fensom and Helen Carter, for assisting with gathering samples from one of the sites investigated and Iben Thomsen and Thomas Kirisits for their advice and suggestions during discussions about our findings. The kind cooperation of Lilburn Estate, John Greenslade, National Forest, Woodland Trust, Salle Farms Estate, Diana MacMullen and Mr and Mrs Barrett is also acknowledged. Conflict of interest statement None declared. Funding Support from Forestry Commission England, Forestry Commission (Corporate and Forestry Services) and Defra (Project TH0119) enabled this study to be carried out. References Baral , H.O. and Bemmann , M. 2014 Hymenoscyphus fraxineus vs. Hymenoscyphus albidus—a comparative light microscopic study on the causal agent of European ash dieback and related foliicolous, stroma-forming species . Mycology 5 , 228 – 290 . Google Scholar CrossRef Search ADS PubMed Chandelier , A. , Helson , M. , Dvorak , M. and Gischer , F. 2014 Detection and quantification of airborne inoculum of Hymenoscyphus pseudoalbidus using real-time PCR assays . Plant Pathol. 63 , 1296 – 1305 . doi:10.1111/ppa.12218 . Google Scholar CrossRef Search ADS Clark , J. and Webber , J. 2017 The ash resource and the response to ash dieback in Britain. In Dieback of European Ash (Fraxinus spp.): Consequences and Guidelines for Sustainable Management . Vasaitis R. and Enderle R. (eds). Swedish University of Agricultural Sciences , pp. 228 – 237 . Downing , E. 2012 House of Commons Library—Ash dieback disease: Chalara fraxinea. http://www.parliament.uk/business/publications/research/briefing-papers/SN06498/ash-dieback-disease-chalara-fraxinea (accessed on 7 January 2016). Drenkhan , R. , Solheim , H. , Bogacheva , A. , Riit , T. , Adamson , K. , Drenkhan , T. , et al. 2017 Hymenoscyphus fraxineus is a leaf pathogen of local Fraxinus species in the Russian Far East . Plant Pathol. 66 , 490 – 500 . doi:10.1111/ppa.12588 . Google Scholar CrossRef Search ADS EPPO 2013 PM 7/117 (1) Hymenoscyphus pseudoalbidus . Bull. OEPP/EPPO Bull 43 , 449 – 461 . CrossRef Search ADS Forestry Commission . 2013 Pest Alert, Ash Dieback Disease. http://www.forestry.gov.uk/pdf/FCPH-ADD.pdf/$file/FCPH-ADD.pdf (accessed on 28 August 2017). Freer-Smith , P.H. and Webber , J.F. 2015 Tree pests and diseases: the threat to biodiversity and ecosystem services . Biodivers. Conserv. 24 . doi:10.1007/s10531-015-1019-0 . Gross , A. and Holdenrieder , O. 2013 On the longevity of Hymenoscyphus pseudoalbidus in petioles of Fraxinus excelsior . For. Pathol 43 , 168 – 170 . Google Scholar CrossRef Search ADS Gross , A. , Holdenrieder , O. , Pautasso , M. , Queloz , V. and Sieber , T.N. 2014 Hymenoscyphus pseudoalbidus, the causal agent of European ash dieback . Molecular Plant Pathol 15 , 5 – 21 . Google Scholar CrossRef Search ADS Gross , A. , Zaffarano , P.L. , Duo , A. and Grünig , C.R. 2012 Reproductive mode and life cycle of the ash dieback pathogen Hymenoscyphus pseudoalbidus . Fungal. Genet. Biol. 49 , 977 – 986 . Google Scholar CrossRef Search ADS PubMed Han , J.G. , Shrestha , B. , Hosoya , T. , Lee , K.H. , Sung , G.H. and Shin , H.D. 2014 First report of the ash dieback pathogen Hymenoscyphus fraxineus in Korea . Mycobiology 42 , 391 – 396 . Google Scholar CrossRef Search ADS PubMed Husson , C. , Cael , O. , Grandjean , J.P. , Nageleisen , L. and Marcais , B. 2012 Occurrence of Hymenoscyphus pseudoalbidus on infected ash logs . Plant Path 61 , 889 – 895 . Google Scholar CrossRef Search ADS Ioos , R. and Fourrier , C. 2011 Validation and accrediation of duplex real-time test for reliable in planta detection of Chalara fraxinea . Bull. OEPP/EPPO Bull 41 , 21 – 26 . Google Scholar CrossRef Search ADS Ioos , R. , Kowalski , T. , Husson , C. and Holdenrieder , O. 2009 Rapid in planta detection of Chalara fraxinea by a real-time PCR assay using a dual-labelled probe . Eu. J. Plant Path 125 , 329 – 335 . Google Scholar CrossRef Search ADS Johansson , S.B.K. , Vasaitis , R. , Ihrmark , K. , Barklund , P. and Stenlid , J. 2010 Detection of Chalara fraxinea from tissue of Fraxinus excelsior using species-specific ITS primers . For. Pathol 40 , 111 – 115 . Google Scholar CrossRef Search ADS Juodvalkis , A. and Vasiliauskas , A. 2002 Lietuvos uosynu dziuvimo apimtys ir jas Iemiantys veiksniai. LZUU Mokslo Darbai [The extent and possible causes of dieback of ash stands in Lithuania] . Biomedicinos Mokslai 56 , 17 – 22 . . Lithuanian with English summary. Keßler , M. , Cech , T. , Brandstetter , M. and Kirisits , T. 2012 Dieback of ash (Fraxinus excelsior and Fraxinus angustifolia) in Eastern Austria: disease development on monitoring plots from 2007 to 2010 . J. Agric. Ext. Rural Dev 4 ( 9 ), 223 – 226 . King , K.M. and Webber , J.F. 2016 Development of a multiplex PCR assay to discriminate native Hymenoscyphus albidus and introduced H. fraxineus and track their changing distribution in Britain . Fungal Biol 23 , 79 – 85 . Kirisits , T. and Freinschlag , C. 2012 Ash dieback caused by Hymenoscyphus pseudoalbidus in a seed plantation of Fraxinus excelsior in Austria . J. Agric. Ext. Rural Dev 4 ( 9 ), 181 – 191 . Kirisits , T. , Kritsch , P. and Krautler , K. 2014 Ascoscarp formation of Hymenoscyphus fraxineus on several-year-old rachises and on woody parts of Fraxinus excelsior. In COST ACTION FP1103 FRAXBACK 6th Management Committee Meeting and Workshop Abstracts, Palanga, LITHUANIA, SEPTEMBER 15–18, 2014, ed Vasaitis, R., pp. 9–10. Kirisits , T. , Kritsch , P. , Krautler , K. , Matlakova , M. and Halmschlager , E. 2012 Ash dieback associated with Hymenoscyphus pseudoalbidus in forest nurseries in Austria . J. Agric. Ext. Rural Dev 4 ( 9 ), 230 – 235 . Kowalski , T. 2001 O zamieraniu jesionów (Dieback of ash) . Trybuna Leśnika Nr 4/359 , 6 – 7 . Kowalski , T. 2006 Chalara fraxinea sp. nov. associated with dieback of ash (Fraxinus excelsior) in Poland . For. Path 36 , 264 – 270 . Google Scholar CrossRef Search ADS Kowalski , T. and Holdenreider , O. 2009 b The teleomorph of Chalara fraxinea, the causal agent of ash dieback . For. Path 39 , 304 – 308 . Google Scholar CrossRef Search ADS Kowalski , T. and Holdenrieder , O. 2009 a Pathogenicity of Chalara fraxinea . For. Pathol 39 , 1 – 7 . Google Scholar CrossRef Search ADS Kowalski , T. and Lukomska , A. 2005 Studies of Fraxinus excelsior L. dieback in stands of Włoszczowa Forest Unit . Acta Agrobotanica 59 , 429 – 440 . Krautler , K. and Kirisits , T. 2012 The ash dieback pathogen Hymenoscyphus pseudoalbidus is associated with leaf symptoms on ash species (Fraxinus spp.) . J. Agric. Ext. Rural Dev 4 ( 9 ), 261 – 265 . Pliūra , A. , Lygis , V. , Suchockas , V. and Bartkevičius , E. 2011 Performance of twenty four European Fraxinus excelsior populations in three Lithuanian progeny trials with a special emphasis on resistance to Chalara fraxinea . Baltic For. 17 , 17 – 34 . PostBOX . 2012 Transmission of Ash Dieback. http://www.parliament.uk/documents/post/Ashdieback_POSTbox.pdf (accessed on 28 August 2017). Przybyl , K. 2002 Causes of Decline of Young Ash Trees and Stands . Central Board of State Forests . (In Polish). Sansford , C.E. 2013 Pest Risk Analysis for Hymenoscyphus pseudoalbidus (anamorph Chalara fraxinea) for the UK and the Republic of Ireland. http://www.fera.defra.gov.uk/plants/plantHealth/pestsDiseases/documents/hymenoscyphusPseudoalbidusPRA.pdf (archived: accessed on 28 August 2017). Schumacher , J. 2011 The general situation regarding ash dieback in Germany and investigations concerning the invasion and distribution strategies of Chalara fraxinea in woody tissue . Bull. OEPP/EPPO Bull 41 , 7 – 10 . Google Scholar CrossRef Search ADS Skovsgaard , J.P. , Wilhelm , G.J. , Thomsen , I.M. , Metzler , B. , Kirisits , T. , Havrdová , L. , et al. 2017 Silvicultural strategies for Fraxinus excelsior in response to dieback caused by Hymenoscyphus fraxineus . Forestry 90 , 455 – 472 . doi:10.1093/forestry/cpx012 . Google Scholar CrossRef Search ADS Thomsen , I.M. and Jorgensen , B.B. 2011 Tilvaeksttab som folge af asketoptorre . Videnblade Skov og Natur 8 , 7 – 50 . (in Danish only). Timmermann , V. , Børja , I. , Hietala , A.M. , Kirisits , T. and Solheim , H. 2011 Ash dieback: pathogen spread and diurnal patterns of ascospore dispersal, with special emphasis on Norway . Bull. OEPP/EPPO Bull 41 , 14 – 20 . Google Scholar CrossRef Search ADS Zhao , Y.-J. , Hosoya , T. , Baral , H.-O. , Hosaka , K. and Kakishima , M. 2013 Hymenoscyphus pseudoalbidus, the correct name for Lambertella albida reported from Japan . Mycotaxon. 122 , 25 – 41 . Google Scholar CrossRef Search ADS Zheng , H.-D. and Zhuang , W.-Y. 2014 Hymenoscyphus albidoides sp. nov. and H. pseudoalbidus from China . Mycol. Progress 13 , 625 – 638 . Google Scholar CrossRef Search ADS © Crown copyright 2018. This article contains public sector information licensed under the Open Government Licence v3.0 (http://www.nationalarchives.gov.uk/doc/open-government-licence/version/3/).
Journal
Forestry: An International Journal Of Forest Research – Oxford University Press
Published: Oct 1, 2018