RABBIT MANAGEMENT FOR GROWERS OF SHORT ROTATION WILLOW COPPICE
Julie Dendy, Gordon McKillop and Richard Watkins. E-mail: [email protected] Central Science Laboratory (CSL)
INTRODUCTION
Short rotation coppice (SRC) is vulnerable to damage by a range of pest species. However, damage by rabbits (Oryctolagus cuniculus) has been identified as one of the main factors which can affect the economic success of the crop, in some cases requiring replanting of the coppice (Beale & Heywood 1997). Rabbit numbers are increasing throughout the country and, given the planned increase in areas of SRC to be grown in the future, the threat rabbits pose to these new plantations is considerable.
This paper provides a synopsis of a recently published advisory leaflet ‘Advice on rabbit management for growers of short rotation willow coppice’ commissioned by the DTI’s New & Renewable Energy Programme (McKillop & Dendy 2000). Its aim is to provide growers and advisors with information on the potential impact of rabbits on SRC plantations and advice on how to manage problem rabbit populations.
RABBIT NUMBERS
The most recent surveys suggest that there may be as many as 30 million rabbits in the UK and numbers are still increasing at a rate of 2% per year (Smith & Trout 1994). Rabbits are estimated to cause around £100 million damage to the UK’s agricultural interests each year, a figure that does not include their impact on forestry or amenity land. As the impact of the lethal virus, myxomatosis, on the population wanes, so numbers will continue to increase. Rabbit numbers are already at pre- myxomatosis levels in some areas, but are, overall, at about 30% of these levels. Accordingly, the potential for damage to SRC plantations is now very high in some areas and increasing in others.
RABBIT BIOLOGY
Adult rabbits weigh between about 1.2-2.0 kg and can eat up to 30% of their body weight in food in a single night (McKillop et al. 1993). Their breeding season extends from January to August, when a succession of litters, usually of 3 to 7 young can be produced by the does at 30 day intervals. Does reach sexual maturity from as early as 4 months old. Despite this prodigious productivity, only 10% of the young will survive longer than twelve months (Cowan & Garson 1985).
The size of a local population, and therefore the risk to a plantation, will be dependent on soil type, climate, the number of predators and the availability of cover. Rabbits are often more numerous on sandy than clay soils and are more abundant in the warmer and drier south-east of the UK. This may be the result of high mortality amongst the young in areas prone to water-logging; the young are particularly vulnerable to the effects of hypothermia (Trout & Smith 1995). Numbers also tend to be higher in areas where predator control is carried out, such as on shooting estates.
1 Warrens are a precious resource for the rabbit as both a refuge and breeding site. Consequently adult rabbits rarely move more than 200m from their burrows and often use well-defined paths or runs within their home range (Cowan et al. 1987). However, juvenile rabbits can disperse, usually at the end of the breeding season, over distances of up to 4km. These individuals do present a threat to SRC, even if the plantation is sited in what was originally a relatively rabbit-free area. Growers should therefore be vigilant to this threat.
RABBIT DAMAGE
Rabbits often damage trees, particularly newly planted ones, by browsing or by ring barking. Browsing is the most common form of feeding damage on young trees and can occur up to a height of 0.5m (higher in snow). It occurs most often in winter and spring and can be identified by a characteristic sharp-angled cut on the end of the stem, with the removed portion often being eaten (Pepper 1998). Ring barking and browsing damage can either kill the young trees or reduce their vigour, making them more susceptible to other pests and diseases or environmental stress (e.g. drought). If the damage is extensive then replanting will be necessary with obvious implications for the profitability of the crop.
ASSESSING THE SIZE OF RABBIT POPULATIONS
Estimates of the distribution and abundance of rabbits in areas to be planted can be obtained by counts carried out at night using a spotlight, or at dawn and dusk under natural light. These techniques only give an index of population size rather than actual numbers present. Nevertheless they provide a useful measure and enables comparisons to be made between sites regarding the potential for rabbit infestation. The best time of year to do this type of monitoring is during the winter months prior to planting, as over-wintering populations at any site are subject to less seasonal variation between years than summer populations. Monitoring of rabbit abundance should be undertaken before and after a control programme to assess its effectiveness and to determine whether further treatments are necessary.
However, growers may not always have time to conduct a series of night-time counts and in newly planted coppices it may be difficult to spot the rabbits amongst the willow shoots. Further work is therefore needed to develop simple census methods that can be conducted quickly and reliably based upon, for example, signs of rabbit occupation such as faecal pellets, burrows or scrapes.
RABBIT MANAGEMENT
If significant numbers of rabbits are present then management programmes need to be carried out to remove rabbits from the proposed SRC site and adjacent land. Rabbit populations, because of the species prodigious reproductive capacity, can withstand high mortalities. Recent work completed in Australia suggests that over 80% of the population should be removed if the grower is to achieve a sustained reduction in rabbit abundance (Twigg et al. 2000).
The most effective time to remove animals is between November and March when natural mortality will have reduced rabbit numbers to their lowest point in the population cycle (Tittensor 1981). Action at this time will reduce the adult breeding
2 population before the next breeding season begins.
A range of lethal and non-lethal methods are currently available to manage rabbit populations (Table 1). Choice of method will vary depending on cost-effectiveness and circumstance. Often growers use several methods in combination, but little information is currently available on the best order in which they should be applied. A number of computer-based pest management support systems are already under development. Landowners or wildlife advisors will be able to feed in information on their pest problem and these systems provide them with a site-specific management programme. The programmes detail the most cost-effective combination of management methods to use, the order in which they should be applied and the optimal time for application. Should information on the costs of rabbit damage to willow plantations become available, similar systems could be developed for SRC plantations; systems that would enable growers to plan and cost rabbit management programmes for their sites.
CURRENT RECOMMENDATION FOR GROWERS OF SRC
Choice of site for SRC plantation
Where possible establish plantations • On heavier soils. • At least 200m away from existing infestations. • Away from woodland. Assessing size of rabbit populations • Estimate rabbit numbers using dawn/dusk or night counts. • Estimate numbers before and after management programmes to determine whether they have been effective.
Rabbit management programmes If significant numbers of rabbits are present • Remove rabbits from plantation site and adjacent land (see Table 1) • Take steps to prevent re-invasion (e.g. protect plantation using wire netting or electric fencing). DISCUSSION Discovering signs of damage or rabbit infestation does not necessarily mean that protective measures need to be taken. The decision should be objectively based upon economic and ecological costs and benefits. However, at present, little research has been conducted to quantify the impact rabbit feeding damage has on yield. Research relating rabbit numbers to extent and costs of damage to SRC is needed along the lines of similar work that has already been conducted for arable farmers. This research showed that rabbit grazing on winter wheat, for example, has a profound effect on crop yields, resulting in losses of about 1%/rabbit/ha which is equivalent to a loss of £6 per rabbit at current yield and grain prices (£70/tonnne).
If this information should become available growers will be able to make informed judgements on: (a) what the likely costs of rabbit damage will be for any one site and consequently whether a control program is required; (b) if control is required, what
3 the most cost-effective management option will be; (c) how long the crop is vulnerable to damage and therefore how long it needs to be protected. For example, it is currently recommended that once rabbits have been removed, new plantations are protected by an effective, but expensive, rabbit-proof fence (MAFF-specification; Pepper 1992). Such a fence provides effective protection for 10 years, but accounts for up to 40% of the plantation’s establishment costs. If the rabbits only have an economically significant impact on the crop during the early stages of its establishment (e.g. the first 18 months), then more temporary and therefore cheaper forms of fencing could be used. We would therefore recommend that damage assessment models, similar to those being developed for arable crops, are also applied to SRC. These models will firstly predict what impact a local population of rabbits will have on yield and, secondly, select the most appropriate and cost-effective option for preventing rabbit damage.
In the longer-term we recommend that future plant breeding programmes should select for rabbit resistance alongside traits for rust and beetle resistance. There is abundant evidence of variation in the susceptibility of different willow species to mammal damage and many of these differences can be associated with the chemical characteristics of the bark. For example, studies in Finland have shown that mountain hares prefer certain willow species over others and mature shoots over juvenile ones (Tahvanainen et al. 1985). There is also evidence to suggest that browsing resistance is greatest in willow species growing in regions subject to severe browsing pressure (Bryant et al. 1989). This evidence indicates that selection for resistance amongst willow species is possible and that trees from regions with high historical browsing pressure could form the focus of future searches for resistant varieties.
The use of rabbit resistant varieties in SRC plantations would reduce the need for costly damage prevention measures by the grower. It would also answer the increasing concerns from the public over the use of lethal control techniques in wildlife management, an important consideration for a developing crop that is reliant on its ‘green’ credentials for public support.
REFERENCES
Beale, C. V. & Heywood, M. J. (1997). Productivity of commercial crops of short rotation coppice at six sites in southern England. Aspects of Applied Biology 49, 181-188. Cowan, D. P. & Garson, P. J. (1985). Variations in the social structure of rabbit populations: causes and demographic consequences. In: Behavioural ecology: the ecological consequences of adaptive behaviour (Eds. R. M. Sibly & R. H. Smith), Blackwell Scientific Publications, Oxford. pp 537-555. Cowan, D. P., Vaughan, J. A. & Christer, W. G. (1987). Bait consumption by the European wild rabbit in southern England. Journal of Wildlife Management 51, 386-392. McKillop I. G., Ginella S. V., Wilson C. J., Hanlon, A. J. & Pugh B. D. (1993). The effects of power failure on European wild rabbits at electric fences. Journal of Applied Animal Behaviour Sciences 35, 277-290. McKillop, I.G. & Dendy, J.A. (2000). Advice on rabbit management for growers of short rotation willow coppice. CSL Advisory Leaflet. 13pp.
4 Pepper, H. W. (1992). Bulletin 102. Forest fencing, Forestry Commission, Edinburgh. Pepper, H. W. (1998). Forestry Commission Practice Note 13. Prevention of mammal damage to trees in woodland. Forestry Commission, Edinburgh. 12pp. Rees, W. A., Ross, J., Cowan, D. P., Tittensor, A. M. & Trout, R. C. (1985) Humane control of rabbits In: Humane Control of Land Mammals and Birds, (Ed. D. P. Britt), UFAW, Potters Bar, London, pp. 96-102. Ross, J. & Sanders, M. F. (1987). Changes in the virulence of myxoma virus strains in Britain. Epidemiology & Infections 98, 113-117. Smith, G. C. & Trout, R. C. (1994). Using Leslie matrices to determine wild rabbit pop growth and the potential for control. Journal of Applied Ecology 31, 223- 230. Tahvanainen, J., Helle, E., Julkunen-Tiitto, R. & Lavola, A. (1985) Phenolic compounds of willow bark as deterrents against feeding by mountain hare. Oecologia 65: 319-323 Tittensor, A. M. (1981). Rabbit population trends in southern England. In: Proceedings of the World Lagomorph Conference (Eds. K. Myers and C. D. McInnes). University of Guelpf, Ontario pp 629-632. Trout, R. C. & Smith, G. C. (1995). The reproductive productivity of the wild rabbit in southern England on sites with different soils. Journal of Zoology 237, 411- 422. Twigg, L.E., Lowe, T.J., Martin, G.R., Wheeler, A.G., Gray, G.S., Griffin, S.L., O’Reilly, C.M. Robinson, D.J. & Hubach, P.H. (2000). Effects of surgically imposed sterility on free-ranging rabbit populations. Journal of Applied Ecology, 37, 16-39.
5 TABLE 1. Cost-effectiveness of main rabbit management techniques. METHOD EFFECTIVENESS MATERIAL LABOUR COST -EFFECTIVENESS OF METHOD (%) COSTS (£, excl. COSTSa (£, excl. VAT) VAT) Wire netting 95+ £1.60/m £1.75/m An extremely cost-effective method of management; designed to give protection at the CSL- (costs of erection one site for up to 10 years; fences need to be maintained on monthly basis to repair specification only; maintenance damage by farm machinery or falling branches and to block any burrows dug under costs are them; in the short term, less cost-effective than electric fencing but little difference in additional) the long term (7+ years) Electric 95+ £1.40/m £0.10/m An extremely cost-effective method of management; designed to be moved from site to fencing (costs of erection site; life expectancy of up to 10 years; fences need to be maintained every few days only; maintenance during first month after erection and thereafter at 2-3 week intervals; in the short term, costs are more cost-effective than wire netting but little difference in the long term (7+ years) additional) Fumigation 90 £0.60/burrow £0.20/burrow The most cost-effective lethal method of management, despite requiring two people to apply it Tree guards 90+ £0.50/tree £0.50/tree More cost-effective than fencing at normal planting spacings when dealing with areas of up to about 2/ha, assuming net guards used Repellents 95 (Aaprotect only) £0.25/tree/year £0.50/tree/year Needs to be reapplied annually; many repellents can give unsatisfactory results; tree guards more cost-effective for small plantations (<2ha), fencing for larger plantations Cage trapping 65 £15/trap £0.50/trap to install The best method for removing rabbits within fenced woodlands; requires visits twice a and £0.50/visit day when set Drop-box 80 (provisional £55/trap £5.00/trap to install Can be used to trap rabbits moving in and out of fenced plantations, although blocking trapping figure) and £0.50/visit burrows under fences and using cage traps is probably more cost-effective; effort needed to site or move traps and is therefore relatively inflexible Ferreting 35 N/A N/A Mainly used because perceived as a sport; time given freely by those undertaking it; not a cost-effective method Shooting 30 £0.10/cartridge N/A Mainly used because perceived as a sport; time given freely by those undertaking it; not a cost-effective method Snares Not recommended - - - Spring traps Not recommended - - - aLabour is costed at £5/hour, the average hourly rate of UK agricultural workers (source, National Farmers’ Union).
6 Integrated Crop Protection in SRC Willow Production
Ming Pei, Tom Hunter and Lori Peacock IACR-Long Ashton Rresearch Station, Department of Agricultural Sciences, University of Bristol, Long Ashton, Bristol BS41 9AF E-mail: [email protected]
Summary
Melampsora rusts and chrysomelid beetles are the most serious disease and pest, respectively, of SRC plantations. Two rust species, M. epitea and M. capraearum, are common in SRC plantations in the UK. They have a complex life-cycle, producing five spore stages and alternating on European larch. In addition to M. epitea and M. capraearum , an asexual form of rust infects young shoots and stems of S. viminalis clones. In M. epitea, there is a large variation in pathogenicity to different willows and three mating populations have been identified. Beetle infestation is often severe locally. Four beetle species occur in willow SRC plantations, the major ones being the blue (Phratora vulgatissima.) and brassy (Phratora vitellinae) willow beetles. Beetle damage is caused firstly by overwintered adults early in the season, followed by larvae and then by newly developed adults late in the season.
Resistance of different willow species/genotypes to rusts and beetles vary greatly. Many willows have consistently shown a high degree of resistance against rust infection and beetle feeding. In some willows, resistance to rust was broken down due to changes in pathogen population. Studies suggest that, compared with monoclonal plantings, mixtures can, in general, reduce rust levels. Against rust, the beneficial effects become particularly apparent with willow clones that show moderate to severe rust infections in monoclonal plantings. Against beetles, blue willow beetle density, damage and, to an extent, number of eggs, tend to be lower in mixtures. Such effects seem to be more obvious as the number of willow clones increases and the distribution of different clones becomes more random. As controlled experiments have shown that the blue willow beetles prefer to feed on S. viminalis, mixtures composed of only S. viminalis clones may not be as effective as mixtures containing different species/species hybrids against beetle damage. Under laboratory conditions, the rust hyperparasite, Sphaerellopsis filum, is capable of reducing willow rust spore production by up to 98%. Recent work has revealed that S. filum varies greatly in pathogenicity to willow rust and its sexual stage occurs in the UK. Studies are now being undertaken to determine how S. filum spread in plantations and how it can be utilised to give best control of the rust disease.
7 Routine use of pesticide in SRC plantations is not practicable for economic, technical and environmental reasons. The desirable control strategies are integrated utilisation of host resistance and natural processes that limit the damage caused by diseases and pests. These include the selection and breeding for resistance, planting genotype mixtures and biological control of rust using S. filum. Willows are hugely diverse and, accordingly, have an immense pool of genes that could be used for disease and pest resistance. In planting mixtures, choosing a range of genetically diverse types is considered to enhance the property of mixtures.
Introduction
As the European Union is legally bound to an 8% reduction of greenhouse gas emission by the year 2010, a large increase of energy from renewable sources is expected. Biomass is one of the major sources of renewable energy and it is likely that, in the near future, increasingly large-scale energy crop plantations will be established. (MAFF, 2000).
Willows (Salix spp.) are the main crop in short rotation coppice (SRC) plantations for renewable energy in the UK and Western Europe. They are grown as the major SRC crop due to their coppicing ability and yield potential. Most SRC plantings in the UK are established with S. viminalis, S. viminalis hybrids with S. caprea and S. cinerea, and S. burjatica..
Like other crops, willows suffer from attacks by diseases and pests. Melampsora rusts and chrysomelid beetles are the most serious disease and pest in SRC plantations. Rust defoliates susceptible plantings prematurely and, when severe may reduce yields by as much as 40% (Parker et al., 1995). Severe rust predisposes plants to infections by secondary pathogens, which often lead to death of the plants. Beetle infestation can be severe, especially at the local scale. The damage by beetles is caused by loss of green leaf area due to foliar feeding, firstly by adults early in the season, followed by larvae and, later in the season, by newly emerged adults. As more and more large- scale plantings become established, disease and pest problem will intensify if no control measures are taken.
Melampsora rusts Two rust species, M. epitea and M. capraearum, are common in willow plantations, with M. epitea being most predominant. Melampsora epitea infects only leaves while M. capraearum infects both leaves and young stems. The two rust species have a complex life-cycle (Fig. 1), overwinter on fallen willow leaves and have an alternate host, European larch (Larix decidua) (Pei et al., 1993).
8 spermatia Larch Willow aeciospore
basidiospores
urediniospore teliospore Fig. 1. Life-cycle of Melampsora epitea and M. capraearum
During late spring – autumn, the rusts are seen as yellow/orange pustules containing urediniospores. The uredinospores are capable of producing the next generation of the same type of spores in 6-7 days. These urediniospores are responsible for many cycles of disease in a growing season resulting in damage on willow. In the autumn, the rusts produce teliospores and overwinter on fallen willow leaves. In spring, the teliospores germinate to produce basidiospores that infect larch. Genetic materials are reshuffled when teliospores germinate and genetic exchange takes place when spermatia fertilise receptive hyphae. Fertilization of receptive hyphae by spermatia triggers formation of aeciospores which infect only willow, not larch.
In addition to M. epitea and M. capraearum, a stem-infecting form (SIF) occurs on S. viminalis genotypes. It has only a single spore stage, overwinters on infected willow stems or buds and causes new infections early in the season (Pei et al., 1995).
Rust variation Within M. epitea, there is large variation in pathogenicity to different biomass willow clones. Many pathotypes (defined according to their ability to infect a certain range of willow genotypes) occur in SRC plantations (Pei et al., 1996). So far 15 pathotypes have been identified from UK sites. Data from pathogenicity tests with rust from UK sites in 1992-1997 suggest that pathotype composition has become increasingly diverse. More pathotypes have been found in South West England and N Ireland where willows have long been grown.
Three distinct mating populations have been found in M. epitea. Each mating population is specific in its pathogenicity to a particular range (section or species) of willows. Until very recently, the likelihood of gene exchange in rust had been determined by difficult and time-consuming crossing experiments. Recent studies suggest that use of molecular tools may have promising prospects in determining genetic relationships between pathotypes.
Rust resistance in willow In the past 10 years, leaf rust (M. epitea) has, in general, sustained similar levels of infection on majority of the genotypes (some 20 – 30) selected in 1980s. However, rust
9 resistance in several high-yielding genotypes, which had been regarded as most promising biomass willows, has been broken down by M. epitea. The main cause of the breakdown is the rise of new virulent pathotypes. An important biomass willow, S. % mollissima ‘Q83’, was highly resistant to rust until 1992 when a new pathotype LET4 arose which rendered this genotype susceptible. DNA fingerprinting of rust samples collected in 1992 revealed that the new pathotype spread from different sources. Salix burjatica ‘Germany’, which was resistant to rust until early 1990’s, has now become one of the most susceptible genotypes. It appears that the present rust population on ‘Germany’ is much more aggressive compared with the rust previously found on the same genotype. The new, highly promising biomass genotype S. viminalis x burjatica ‘Stott 10’ can yield twice as much as most biomass genotypes selected in 1980s. It has been highly resistant to rust in most UK sites since it was bred in late 1980s. However, increased levels of rust on this genotype have been found at a site in Northern Ireland and a site in Southern England since 1997. The willow genotypes introduced or developed in early 1990s as a result of breeding (about a dozen) were mostly less susceptible to leaf rust (slight increase of leaf rust infection has been observed on some of these genotypes in recent years).
The stem-infecting form (SIF) occurs specifically on S. viminalis clones. Most of the commercial biomass willows currently available in UK and European markets are S. viminalis clones and its hybrids with S. schwerinii, produced by Svalof Breeding programme in Sweden and introduced into the UK in early 1990s. While these clones have generally been resistant to leaf rust, several are infected by SIF in the UK.
Results from inoculation tests and field disease assessments indicate that majority of willows introduced from the Far East and North America are resistant to UK rust populations.
Chrysomelid beetles The main pests of willow plantations are members of Chrysomelidae. Phratora vulgatissima (blue willow beetle) and P. vitellinae (brassy willow beetle) are the most prevalent while, occasionally, damage is caused by Galerucella lineola and Plagiodera versicolora. The life-cycle of P. vulgatissima is shown in Figure 2. In April, adults emerge from overwintering sites (usually outside the plantations) to find willow plantings on which to feed. Female beetles lay eggs on willow leaves and the resulting larvae are responsible for further damage in the plantation during the summer months. In late summer and early autumn, new adult beetles emerge from the pupae developed from the larvae. After approximately 6-8 weeks, these beetles will leave and hibernate in suitable overwintering sites such as bark crevices and old fence posts. It has been suggested that in some years the other willow beetles may produce a second generation of adults.
Summer
Larval feeding damage
Eggs Larvae Pupae on willow on willow in soil Adult feeding damage Adults new generation of adults emerging
Adults leave overwintering Adult feeding damage Spring sites and fly to willows Autumn where they feed, mate and begin egg laying Adults leave willows and disperse 10 (fly) to overwintering sites
Adults Fig. 2. Life-cycle of blue willow beetle Phratora vulgatissima
Feeding damage in willow To assess the effect of beetle infestation on biomass yields, beetle feeding damage was simulated by removing leaves from growing plants of S. viminalis ‘Bowles Hybrid’ (Fig. 3., from Kendall & Wiltshire, 1998). In the most extreme case, where leaves were removed in both June and August, a defoliation of 90% resulted in approximately 40% yield loss. Previous field surveys have shown that, in general, total leaf area loss by beetle feeding has been in the region of 10 – 30% (Hunter and Peacock, personal observations). A loss of 10 - 30% of the total leaf area resulted in approximately 8 - 22% yield loss in the simulated defoliation experiments. However, if damage is substantial over a number of years, the consequence will be more severe.
100 95 90 85 80 75 70 early (June)
dry weight (% of control) 65 late (August) early + late 60 55 0 102030405060708090100 % leaves removed
Fig. 3. Effects of simulated defoliation on biomass yields (means of ANOVA) of S. viminalis ‘Bowles hybrid’. From Kendal & Wiltshire. 1998.
Beetle feeding preferences Surveys of willow plantations indicate that the severity of chrysomelid beetle infestation varies between years and between locations (Hunter et al., 1996). Laboratory and field studies have shown that the willow beetles exhibit feeding preference differences between willow genotypes (Wiltshire et al., 1997). Various
11 chemicals, including phenolic glucosides, have been found in willow leaves. In general, S. viminalis and its hybrids are favoured by both P. vulgatissima and G. lineola. It appears that P. vulgatissima prefers willows with low levels of phenolic glucosides while P. vitellinae prefers willows with high levels of these chemicals.
Disease and pest control
SRC willow is a new crop and there are no long-established protocols for disease and pest control. Unlike most food and horticultural crops, routine use of fungicides and pesticides in SRC plantations is not considered to be a viable option because (a) it is too costly for the low-input crop, (b) it is difficult to obtain effective coverage of chemical over the large and dense canopy and (c) application of large quantities of agrochemicals may have harmful effects on the environment. A desirable strategy of disease and pest control in SRC plantations is the integrated utilisation of host resistance and natural processes that limit the damage by diseases and pests.
HOST RESISTANCE Willows are hugely diverse (300-500 species, each species having numerous genotypes), hybridise with relative ease and only need 1-3 years to reproduce. Studies have shown that a wide range of willows, particularly those introduced from the Far East and North America, are resistant to rust and beetles. These attributes make the use of natural resistance one of the most attractive options to control diseases and pests. Currently, two willow breeding programmes, one based at Long Ashton, UK, and one at Uppsala, Sweden, are underway. These programmes aim at producing disease and pest resistant, high yielding genotypes and have already produced hundreds of crosses for selection and evaluation.
As rusts are specialised fungi, changes or shifts of pathogenicity in the population may result in severe damage to previously resistant crops. In the past 10 years, rust resistance in several high-yielding genotypes, which had been regarded as amongst the most promising biomass willows in the UK, has been broken down by M. epitea. Therefore, breeding for durable resistance is particularly important in deploying genetic resistance. An understanding of the behaviour of pathogen populations and breeding based on broad genetic backgrounds will help to produce novel genotypes having long- lasting resistance against rust.
Willow mixtures Large-scale monoculture plantings are known to be vulnerable to attacks by diseases and pests. A strategy to reduce such risks is to plant genotype mixtures to increase genetic diversity. Because genotype mixtures are readily accepted for biomass production, willows have a unique advantage over food or horticultural crops, in which flavour, colour or shape may have prime importance. The experience in the UK over the last 10 years suggests that, compared with monoclonal plantings, mixtures
12 can, in general, reduce disease and pest levels. In particular, the beneficial effects of mixtures against rust become apparent with the genotypes which show moderate to severe rust infections in monoclonal plantings (Parker et al., 1995; McCracken & Dawson, 1998). Studies have shown that the planting of willow mixtures may also reduce the extent of damage by beetles. Compared with monocultures, beetle density, damage and, to an extent, number of eggs, tend to be lower in mixtures containing willows that differ in their susceptibility to beetles (Peacock & Herrick, in press). Such effects seem to be more obvious as the number of willow clones increase and the distribution of clones becomes more random. Currently, many SRC plantings are mixtures containing only S. viminalis clones. As experiments have shown that beetles prefer to feed on S. viminalis, S. viminalis single species mixtures may not be as effective as mixtures containing different species/species hybrids against beetle damage. One of the major concerns over planting genotype mixtures is that, because virulence patterns in pathogens can change quickly through gene exchange during the sexual life-cycle, the mixtures may favour the emergence of ‘super races’ that are capable of attacking all components of mixtures. If that is so, the effect of the mixtures would be greatly reduced (Wolfe, 1985). Our studies suggest that M. epitea consists of several different mating populations (Pei et al., 1999a, b). For example, the pathotypes occurring on S. viminalis genotypes do not exchange genes freely with those on S. burjatica. By choosing genotypes that harbour genetically separate pathogen populations, the property of mixtures to suppress rust would be enhanced. At present, because of availability, most new SRC plantings are mixtures containing only S. viminalis genotypes plus one or two S. viminalis hybrids. Whether such mixtures encourage the pathoptypes having wide spectra of virulence is an important question to be answered in the near future.
In planting willow mixtures, certain characteristics, such as growth habits, soil and water conditions, and adaptability of the willows to the environment, need further consideration.
Biological control of rust Compared with annual crops, SRC is better suited for biocontrol because of the carry- over effect on biological control agents, assuming a 3-5 year harvest interval. Sphaerellopsis filum is a fungal hyperparasite. It attacks a wide range of rust fungi, including willow Melampsora. Under experimental conditions, the hyperparasite can reduce willow rust spore production by up to 98%. Although S. filum occurs only on rusts in nature, it can be cultured on artificial media. This provides the opportunity of growing the fungus in vitro and then reintroducing it into the field.
Previously, only the asexual stage of S. filum was known in the British Isles. Asexual conidiospores are waterborne and dispersed mainly by rain splash. Recently, it was found that sexual ascospores occur in S. filum in the UK (Yuan et al., 1998). Ascospores are airborne and, hence, are more efficiently dispersed. Until very recently, it was not clear whether S. filum is specialised in its pathogenicity. Recent studies revealed that S. filum varies in pathogenicity and some strains are significantly more virulent than others to willow Melampsora (Yuan et al., 1999). Answers to the following will help to deploy S. filum more effectively for biocontrol of rust in SRC
13 willow: (a) to what extend S. filum varies in pathogenicity, (b) whether S. filum produces the sexual stage on willow rust and (c) how it spreads.
Other measures As European larch serves as an alternate host of willow rust, planting SRC willows near larch will cause early disease onset and more severe infection. Therefore, willow plantings should be sited as far away from larch plantations as practicable to delay the onset and reduce rust severity.
When fungicides were used for studies of effects of rust disease on growth and yields, it was found that myclobutanil (Systhane) and benodanil (Calirus) are effective against rust (Royle, 1991, McCracken & Dawson, 1998). Field experiments showed that, on susceptible genotypes, at least 5 applications were needed to maintain low levels of rust throughout the season (Royle, 1991).
Conclusion
The potential threat of diseases and pests on SRC willow is likely to increase as more and more willow plantings are established. Integrated control of SRC willow diseases and pests involves the breeding for resistant genotypes, planting host genotype mixtures and deployment of biological control agents. The encouraging prospect is that these integrated approaches can benefit from the huge diversity in willows, acceptance of mixtures for energy production and the perennial nature of the crop.
Acknowledgement
We thank the Ministry of Agriculture, Food and Fisheries (MAFF) and the DTI’s New & Renewable Energy Programme (managed by ETSU), UK, and the European Commission for financing the work. IACR receives grant-aided support from the Biotechnology and Biological Sciences Research Council of the United Kingdom.
References
Hunter, T.; Royle, D.J. ; Arnold G.M. (1996). Variation, in the occurrence of rust (Melampsora spp.) and other diseases and pests, in short rotation coppice plantations of Salix in the British Isles. Annals of Applied Biology 129; 001-012.
Kendall, D. A. ; Wiltshire. C. W. (1998) Life-cycle and ecology of willow beetles on Salix viminalis in England. European Journal of Forest Pathology 28, 281-288.
MAFF (2000) England Rural Development Plan: Energy Crop Scheme Consultation Document. Available from URL:www.maff.gov.uk/farm/acu/acu.htm
14 McCracken, A. R.; Dawson, W. M. (1998) Short rotation coppice willow in Northern Ireland since 1973: development and use of mixtures in the control of foliar rust (Melampsora spp.). European Journal of Forest Pathology 28, 241-250.
Parker, S. R.; Pei, M. H.; Royle, D.J., Hunter, T.; Whelan, M.J. (1995) Epidemiology, population dynamics and management of rust diseases in willow energy plantations. Final Report of Project ETSU B/W6/00214/REP. Energy Technology Support Group, Department of Trade and Industry, UK.
Peacock, L.; Herrick, S. Responses of the willow beetle Phratora vulgatissima to genetically and spatially diverse Salix spp. plantations, Journal of Applied Biology (in press).
Pei, M.H.; Royle, D. J.; Hunter, T. (1993) Identity and host alternation of some willow rusts (Melampsora spp.) in England. Mycological Research 97, 845-851.
Pei, M.H.; Royle, D. J.; Hunter, T. (1995) A comparative study of stem-infecting and leaf-infecting forms of Melampsora rust on Salix viminalis in the UK. Mycological Research 99, 357-363.
Pei, M. H.; Royle, D. J.; Hunter, T. (1996) Pathogenic specialization in Melampsora epitea var. epitea on Salix. Plant Pathology 45, 679-690.
Pei, M. H.; Hunter, T.; Ruiz, C. (1999a) Occurrence of Melampsora rusts in biomass willow plantations for renewable energy in the United Kingdom. Biomass and Bioenergy 17, 153-163.
Pei, M. H.; Hunter, T.; Royle, D. J. (1999b) Hybridisation in larch-alternating Melampsora epitea (M. larici-epitea). Mycological Research 103, 1440-1446.
Royle, D. J. (1991) Potential Pest and disease problems associated with arable energy crops. In Wood Fuel for Thought, Richards, G. E. (Ed), Harwell Laboratories, Oxford, pp 283-294.
Wiltshire C. W.; Kendall D. A.; Hunter T.; Arnold G.M. (1997) Host-plant preferences of two willow-feeding leaf beetles (Coleoptera, Chrysomelidae) Aspects of Applied Biology 49, 113-120.
Wolfe, M. S. (1985) The current status and prospects of multiline cultivars and variety mixtures for disease resistance. Annual Review of Phytopathology 23, 251- 273.
Yuan, Z. W.; Pei, M. H.; Hunter, T.; Royle, D. J. (1998) Eudarluca caricis, the teleomorph of the mycoparasite Sphaerellopsis filum, on blackberry rust Phragmidium violaceum. Mycological Research 102, 866-868.
Yuan, Z. W.; Pei, M. H.; Hunter, T.; Ruiz, C.; Royle, D.J. (1999) Pathogenicity to willow rust, Melampsora epitea, of the mycoparasite Sphaerellopsis filum from different sources. Mycological Research 103, 509-512.
15 Willow Aphids on SRC Willow Effects of aphids on host plants: A project summary
C. Matilda Collins, Department of Biology, Imperial College of Science Technology and Medicine, Silwood Park, SL5 7PY. Tel: 020 7594 2420 Fax: 020 7594 2308 e-mail: [email protected]
ABSTRACT Tuberolachnus salignus and Pterocomma salicis are both aggregative aphids that have been noted within SRC crops. Experiments with willows grown in soil and in hydroponic culture reveal that T. salignus can reduce the above-ground yield of biomass willows, have severe negative effects on the roots and reduce the survival of both newly planted and established trees. Pterocomma salicis can also reduce yield and negatively affect the roots of host trees, but its effects are less marked. Both species can lead to plant stress and increase the vulnerability of host trees to other factors. Field observations indicate that T. salignus may be increasing in abundance in SRC crops.
INTRODUCTION Insect pests can have adverse impacts on SRC yield and thus on the economics of production. Previous entomological studies have concentrated on the more obvious damage to SRC caused by defoliating insects. The damage caused by both leaf and stem-feeding aphids is largely invisible; other than in their roles as viral vectors and gall inducers they rarely cause obvious alterations to leaf structure and function. However, aphid presence is an energy and nutrient drain due to their phloem-feeding habit. The energy and nutrient loss caused to sycamore trees (Acer pseudoplatanus) by the sycamore aphid (Drepanosiphum platanoidis) has been shown to lead to earlier leaf fall, to reduce the size of leaves by 40% and the production of stem wood by 62% (Dixon, 1971). The sycamore aphid is much smaller than the giant willow aphid, Tuberolachnus salignus, which has been shown to ingest the photosynthetic product of 5-20 cm2 of leaf per day (Mittler, 1958). The magnitude of this drain has the potential to affect significantly the yield of willows by altering the source-sink relationship within the plant. There may also be potential impacts on the long-term development and fitness of the rootstock left in situ after coppice harvesting as a result of changes in root health, density and architecture.
THE APHIDS
This study has focussed on two aphid species: Tuberolachnus salignus, the giant willow aphid and Pterocomma salicis, the black willow aphid. Both species aggregate and occur in dense colonies on the stems of infested plants, but have contrasting ecologies.
Tuberolachnus salignus (Gmelin), Lachninae, Lachnini The Giant Willow-aphid is one of the largest aphids ever recorded (BL 5.0-5.8 mm) (Blackman & Eastop, 1994). It feeds almost exclusively on willow, but has very occasionally been recorded on poplar (Populus sp.). Its distribution reflects that of willows; it is virtually cosmopolitan, only being absent from Australasia. The species is apparently parthenogenetic; no sexual morphs have ever been found and recent DNA studies support this (Blackman & Spence, 1996). The species is strongly aggregative, forming vast colonies on infested trees. These colonies can cover much of the 1-3 year old stem surface of a tree. Other than its large size the most distinctive features of this aphid are a large dorsal tubercule (function unknown), the reddish stain it makes when squashed and synchronised waving of the hind legs over large areas of infestation. Many entomological texts note this insect as being capable of much damage, even of killing willows up to 40 feet tall (Buckton, 1881; Das, 1918; Swirski, 1963), but until now little attention has been paid to its ecology and host relationships.
Pterocomma salicis (Linaeus), Aphidinae, Pterocommatini The Black willow-aphid is another large aggregative aphid (BL 2.7-4.5 mm). It also feeds exclusively on willow (Blackman & Eastop, 1994), and is common in the northern hemisphere. Although the species is parthenogenetic for much of the year, a sexual generation is produced in autumn and the
16 overwintering egg is found in the buds and on the stems of willows. The species is often attended by ants, which may protect it from predation.
In this report I summarise studies on the impact of these two species on the establishment and yield of SRC willows.
MATERIALS AND METHODS For full details please refer to my PhD Thesis (Collins, 2001). Field monitoring. Aphid populations on four willow varieties have now been monitored for three years at Roves Farm, Sevenhampton, Wiltshire, U.K. (Grid Reference: SU 210 888). 20 coppice stools of ‘Bowles Hybrid’, and 21 stools each of ‘Jorr’, ‘Orm’ and ‘Dasyclados’ were assessed for aphids weekly in 1998 and fortnightly in 1999 and 2000. Effects of aphids on host plants Willows grown in soil: Fifty-two potted cuttings of the hybrid willow clone ‘Jorr’ were randomly assigned to one of four treatments: 1 No aphids 2 Inoculation with 20 Pterocomma salicis on 12th April 1999 3 Inoculation with 5 Tuberolachnus salignus on 25th June 1999 4 Inoculation with both the above. The aphid infestations were monitored weekly and assigned to levels in a six point scale: 0.<5 aphids, 1. 5 – 20 aphids, 2. 20 – 50 aphids, 3. 50 – 100 aphids, 4. 100 – 300 aphids, 5. >300 aphids The levels recorded on a weekly basis were summed at the end of the experiment to estimate the total level of infestation to which each tree was subject. Harvesting. The first harvest of above ground biomass took place nineteen weeks after infestation with P. salicis and nine weeks after infestation with T. salignus. The trees were subsequently maintained ‘aphid free’ and allowed to recover. The second harvest took place ten weeks later. Post harvest recovery. After the first harvest, trees were assessed weekly against a phenology scale in order to provide an assessment of the speed of their recovery from coppicing. Willows grown in water cultures to give greater detail: Tuberolachnus salignus. These experiments were carried out in a controlled environment chamber set at 15(±1)°C and 16 hours of light to 8 of darkness. In the first experiment the willows were infested at planting, in the second the cuttings were allowed 60 days to establish prior to infestation. Aphid inoculation was with a single young adult alatae. Harvesting took place 25 days after infestation in both cases. Pterocomma salicis. This species is less amenable than T. salignus and insufficient numbers established under controlled environment conditions. A subsequent repetition of the experiment in a glasshouse (min temp 15±1°C, ambient June daylight regime) has yielded data for infestation at planting, but not for infestation after 60 days establishment. Measurements: For each plant, the number of leaves was counted weekly, this served as a non- invasive growth measure. In the first experiment the number of primary roots emerging were also counted; the longer duration of the second experiment made this impossible and roots were measured destructively at harvest. Wet and dry mass of the roots, shoots and twig were determined after harvest. Photosynthetic rate: 12 pot grown ‘Dasyclados’ individuals were randomly allocated to one of two treatments: control and with aphids. Trees were infested by inoculation with 5 4th instar T. salignus apterae and populations were left to develop for three weeks prior to sampling. Photosynthetic rate was determined using an IRGA. 12 measurements were made per tree, four at each of three sites: the branch tip, the mid-stem and the base. Nitrogen concentration: For analysis of leaf nitrogen content, five leaves were randomly selected from each tree used in the photosynthesis experiment. The leaves were oven dried for 48 hours at 80°C, ground in a ball mill to homogenise and a sub-sample of this digested in sulphuric acid and selenium at 400°C for 1.5 hours (Allen et al., 1989). Total nitrogen content was determined using a Kjeltec nitrogen analyser (Persop Analytical, Bristol) and the % dry mass calculated.
17 RESULTS Aphids in the field Over the last three years the giant willow aphid has increased in abundance in Jump Field. In October 1998 there was a peak number of 14,000 aphids on the sample of 83 trees, in October 1999 this figure was over 190,000 (Figure 1). Not only have the numbers of aphids increased, but also the duration of infestation. The increase in duration of infestation, from 10.24 ±0.74 weeks in 1998 to 17.17 ±0.54 weeks in 1999 is highly significant (T=7.52, d.f.=164, p<0.001). When the survey work for 2000/2001 is completed we will be able to see if this trend of increased abundance is continued. This species has an unusual temporal pattern of occurrence for a tree-feeding aphid. It increases in summer, normally considered a time of poor host quality and remains at high levels until the early spring, when it vanishes from the sample for several months – a portion of the lifecycle that remains unexplained. Its continued presence on trees that have shed their leaves, and are thus not photosynthetically active, indicates unusual feeding abilities. Additionally a high degree of cold tolerance is needed to endure winter frosts and there are indications that survival is high at temperatures as low as –5°C (L. Peacock, Pers. Comm.). No predators have been observed regularly feeding on this species and there has also been no evidence of parasitoid activity. In the winters of 1998 and 1999 colonies of giant willow aphid were decimated by an entomopathogen; the aphids swell and die, but remain hanging by their stylets for several days before exploding. It has not yet been determined whether this pathogen is of viral or fungal origin, but after three years of observation this entomopathogen remains the only prospect for biological control of the giant willow aphid. The giant willow aphid excretes a substantial amount of sugar rich honeydew, which attracts large numbers of other insects into the crop, possibly influencing the species diversity of willow stands. This accessible carbon source may also have a profound influence on soil chemistry and further studies are in progress to examine this area (Collins & Crawley, Submitted) The black willow aphid population in this crop has declined since the peak numbers of this project were recorded in May 1998, and relatively few were found in 1999. The duration of infestation also declined significantly between 1998 and 1999 (T=5.47, d.f.=164, p<0.001) from 2.61 ±0.31 weeks in 1998 to 0.63 ±0.18 weeks in 1999. There is anecdotal evidence from the landowner, Mr Rupert Burr, that P. salicis reached a peak of abundance in 1997, prior to this sampling programme. In both 1999 and 2000 this species has largely been found in mixed-species colonies with T. salignus, but their numbers now appear to be rising again. This species is consumed by many of the common aphid predators; hoverfly, lacewing and ladybird larvae have all been observed feeding on the black willow aphid. A high level of parasitioid activity was also observed coincident to the decline in numbers in the summer of 1998, with hyperparasitoids and mummy parasitoids also being very abundant (F.van Veen, pers. comm.). Effects of aphid infestation on yield. In all aphid treatments the biomass at first harvest was reduced compared to the control treatment (F=42.87, d.f.=3,48, p<0.001)(Table 1). Tuberolachnus salignus, which reduced host biomass to less than half that of the controls in nine weeks of infestation, has a greater impact on yield than does Pterocomma salicis. The presence of both species on the same tree does not cause an additive decline in tree performance, indeed, the black willow aphid appears both to reduce the numbers of the giant willow aphid and to mitigate its impact. Infestation with T. salignus, whether alone or with P. salicis, delayed host recovery after harvest (Figure 2). Infestation with P. salicis alone had no impact speed of recovery. The presence of T. salignus prior to first harvest continues to have a negative impact on the host tree at second harvest (F=6.15, d.f.=3,38, p<0.002)(Table 1). As aphids were not present during the inter- harvest period, this implies a long-term impact on the growth of the host plant. The proportion of dry matter in the shoots at second harvest is higher in the T. salignus treatment than in the others. In spite of the absence of direct aphid influence, this growth is not only quantitatively different from the control but also qualitatively. A long-term effect such as this is likely to be mediated by the roots. Intriguingly, the trees that had been host to the black willow aphid were larger than the controls at second harvest. This could however be an artefact of the restricted rooting area available in the pots; the large control trees may have become pot-bound prior to first harvest, thus restricting their biomass increase between the two harvests. Trees infested by P. salicis grew less in the first stage and may have had more opportunity for root growth, and thus shoot growth, in the second stage. Aquaculture – the details Tuberolachnus salignus growth rates: Although only a single alate was used for inoculation, under these experimental conditions an aphid clone can rapidly increase in biomass. Alates have a mean (± 1SE) mass of 5.28 (± 0.17)mg, give birth to 34.3 (± 1.4) nymphs which in turn can reach reproduction
18 within 28.11 (± 0.25) days. Thus, within the 25 days of each experiment each alate can lead to an aphid biomass per twig of 346.7 (± 17.8) mg (Collins & Leather, In Press). Infested at planting. The performance of the willows in each treatment diverges immediately. The pattern of above ground growth in this experiment is similar to the results from soil grown trees: both species had a severe effect on host plant growth. There is also a qualitative difference in the above ground growth. Aphid infested shoots have a significantly lower water content than the control shoots, although the effect of P. salicis is minor. By the first root and leaf count on the 11th day of the experiment aphid infested trees have a significantly lower number of primary roots and leaves (roots: t=3.32, df=26 p<0.01, leaves: t=3.47, df=26, p<0.001). In the infested treatment some of the plants were observed to reabsorb previously initiated roots. The dry mass of roots is significantly lower in both aphid treatments than in the controls, with T. salignus having the greater impact. Additionally roots in the T. salignus infested treatment had a higher proportion of water than control roots (Table 2). The aphids also have an impact on the woody tissue. Significantly less of the original twig mass remains in the aphid infested treatment. Infested at 60 days. There was no visual difference in size or significant difference in number of leaves between treatments at infestation (t=0.80, df=31, p=0.42). Ten days after infestation the mean number of leaves per plant in each treatment differs (t=4.46, df=31, p<0.001). Giant willow aphid infested trees shed their lower leaves and apical growth was vastly reduced. In spite of the reduced number of leaves, the water consumption per plant rises markedly after infestation, then falls as the plant starts to weaken and wilt (Figure 3). As above the biomass of shoots and roots are significantly reduced in the aphid treatment compared to the control. Again, there is also a qualitative difference in both the above and below ground growth. Aphid infested shoots were dehydrated compared to control shoots and roots of infested trees had a much higher water content than control roots. Photosynthesis and leaf nitrogen content. Both of these measures were significantly raised by the presence of Tuberolachnus salignus. The photosynthetic rate was raised in all three positions on the tree, i.e. above and below the level of aphid infestation (Table 3). Photosynthesis is a water-demanding plant metabolic process and with reduced root capacity, this may be a key mechanism in the plant stress induced by aphid presence.
CONCLUSIONS Both these aphid species do adversely impact the yield of SRC willows. The black willow aphid’s effect is less marked than that of the giant willow aphid, which can reduce not only the growth the plants make subsequent to infestation but also alter previously developed woody tissue. The negative effects of the giant willow aphid on the shoots and roots of established and establishing trees are drastic and are both quantitative and qualitative. Impoverished roots can reduce the survival of infested trees as well as increasing the vulnerability of the host to other potentially detrimental factors. It is likely that mortality previously attributed to drought or frost may have been exacerbated or increased by the presence of these stem-feeding aphids. As yet we understand little of the population dynamics of these aphids and long-term monitoring within extensive SRC crops is required to determine the frequency and abundance of their occurrence. A greater understanding of the ecology of these species is also required to enable the investigation of natural enemies and entomopathogens as potential control measures.
ACKNOWLEDGEMENTS I wish to thank my supervisors Simon Leather and Rufus Sage, the Game Conservancy Trust and the DTI’s New & Renewable Energy Programme (managed by ETSU) for their financial support and Rupert and Joanna Burr for allowing me unrestricted access to their willow crops.
REFERENCES Allen, S.E., Grimshaw, H.M., Parkinson, J.A., & Quarmb, C. (1989) Chemical Analysis of Ecological Materials Blackwells, Oxford. Blackman, R.L. & Eastop, V.F. (1994) Aphids on the world's trees: an identification and information guide CAB International in association with The Natural History Museum.
19 Blackman, R.L. & Spence, J.M. (1996) Ribosomal DNA is frequently concentrated on only one X chromosome in permanently apomictic aphids, but this does not inhibit male determination. Chromosome Research, 4, 314-320. Buckton, G.B. (1881) Monograph of the British Aphides The Ray Society, London. Collins, C.M. (2001) Aspects of the ecology of two stem-feeding willow aphid species. PhD, Imperial College of Science Technology and Medecine. Collins, C.M. & Crawley, M.J. (Submitted) Indirect effects of aphid feeding: an experiment with artificial honeydew. Functional Ecology. Collins, C.M. & Leather, S.R. (In Press) The effect of temperature on the fecundity and development of the Giant Willow Aphid, Tuberolachnus salignus (Gmelin) (Lachninae, Lachnini). European Journal of Entomology. Das, B. (1918) The Aphididae of Lahore. Memoirs of the Indian Museum, 6, 135-274. Dixon, A.F.G. (1971) The role of aphids in wood formation. I. The effect of the sycamore aphid Drapanosiphum platanoidis (Schr.) (Aphididae), on the growth of sycamore, Acer pseudoplatanus (L.). Journal of Applied Ecology, 8, 165-179. Mittler, T.E. (1958) Studies on the feeding and nutrition of Tuberolachnus salignus . II.-The Nitrogen and sugar composition of ingested phloem sap and excreted honeydew. Journal of Experimental Biology, 35, 74-84. Swirski, E. (1963) Notes on Plant Lice (Aphidoidea) of Israel. Israel Journal of Agricultural Research, 13, 9-23.
20 Figure 1. The total number of Tuberolachnus salignus (solid line) and Pterocomma salicis (dotted line) present on 83 willow coppice stools at Jump Field, Roves Farm (G.R. SU 210 888) from February 1998 to September 2000. Note the logarithmic response axis.
Figure 2: The individual and combined effect of previous infestation with the aphids Tuberolachnus salignus and Pterocomma salicis on the speed of recovery of coppiced shoots of the willow clone ‘Jorr’. The response axis represents a phenological scale against which recovery was assessed, 1 represents swelling of the buds, 6 full leaf development.
Figure 3: The effect of Tuberolachnus salignus on water use (ml/day/tree) by saplings of the SRC willow clone ‘Jorr’ grown in water cultures
21 Table 1: The effect of the aphids Pterocomma salicis and Tuberolachnus salignus compared to controls on the above-ground biomass, losses and water content of saplings of the SRC willow clone ‘Jorr’ grown in soil. All values given are significant at 5%. n.s. denotes non-significant differences.
Aphid Species: Pterocomma Tuberolachnus Both Species salicis salignus Shoot dry mass at 1st harvest - 23% - 52% - 47% losses to 2nd harvest 8 % 33 % 47 % Shoot dry mass at 2nd harvest + 95% - 67% - 54% water content of shoot mass n. s. - 36% n. s.
Table 2: The effect of the aphids Pterocomma salicis and Tuberolachnus salignus compared to controls on the biomass and water content of shoots, roots and woody tissue of saplings of the SRC willow clone ‘Jorr’ grown hydroponically. All values given are significant at 5%. n.s. denotes non- significant differences, - that data was not available.
Aphid Species: Pterocomma Tuberolachnus salicis salignus Infested at planting Shoot dry mass - 53% - 58% Shoot water content - 2% - 66%
Root dry mass - 53% - 84% Root water content n. s. + 68%
woody tissue dry mass n. s. - 5% Infested at 60 days Shoot dry mass - - 30% Shoot water content - - 19%
Root dry mass - - 50% Root water content - + 44%
woody tissue dry mass - - 10%
Table 3: The effect of Tuberolachnus salignus on the net rate of photosynthesis (µmol/m2/s) and nitrogen content (% dry weight) of leaves of the SRC willow clone ‘Dasyclados’. All values given are significant at 1%.
Net photosynthetic rate % N Base Mid Tip + 119% + 80% + 76% +22%
22 1000000 Tuberolachnus salignus Pterocomma salicis
100000
10000
1000
100
10
1 Feb-98 May-98 Aug-98 Nov-98 Feb-99 May-99 Aug-99 Nov-99 Feb-00 May-00 Aug-00
23 40
35
30 Control Tuberolachnus salignus 25
20
ml/plant/day 15
10
5
0 0 102030405060708090 Time in Days
24 IMPROVING BREEDING EFFICIENCY OF BIOMASS WILLOWS USING MOLECULAR MARKER TECHNOLOGY
S J Hanley, J H A Barker and A Karp
IACR- Long Ashton Research Station, Department of Agricultural Sciences, University of Bristol, Long Ashton, Bristol BS41 9AF, UK E-mail: [email protected]
Abstract
Increased energy production from renewable sources is currently a high priority in Europe. At present, biomass accounts for 3% of European inland energy consumption. It has been proposed that by 2010 this figure should be increased to 8% or 90 million tonnes of oil equivalent (mtoe). To meet this target, efforts are underway to breed and produce higher yielding biomass cultivars. While significant advances have been made in this area, the efficient coupling of conventional breeding methods and molecular marker technologies are required to maximise yield increases and provide sustainable disease and pest resistance for the future. Molecular marker technologies can accelerate the efficiency of breeding programs by helping to locate genes controlling agronomic traits and provide markers for marker-assisted selection. In order to achieve this, genetic maps with very large numbers of markers have to be produced to increase the chance of finding a marker closely linked to a trait of interest. The technique of Amplified Fragment Length Polymorphism (AFLP) generates large numbers of markers in a single assay but these markers cannot easily be transferred from one cross to another. Microsatellites provide fewer markers but can be used to anchor maps from different crosses. Towards marker-assisted selection in willows, a genetic map of a full-sib cross of Salix viminalis is being constructed. In addition, two reference families, each comprising 950 progeny, have been established and initial trait assessments, including laboratory-based beetle feeding assays and rust pathogenicity tests, have been performed. To identify markers linked to traits of interest, two approaches will be used:
• QTL analysis - associations between markers and traits will be identified using software packages that identify quantitative trait loci (QTL)
• Bulked Segregant Analysis (BSA) – DNA from individuals of contrasting phenotype (e.g. resistant and susceptible) will be bulked and markers which correspond to only one phenotype will be identified
Furthermore, in an attempt to merge the mapping efforts of willow with poplar, the efficacy of using poplar microsatellite markers for mapping studies in willow has been demonstrated.
Introduction
25 A current limitation facing the breeding of improved willow varieties is our lack of knowledge on the genetic basis of important agronomic traits. In this project we are applying state-of-the-art molecular marker technologies to gain an insight into these fundamental issues. In many crops, it has been demonstrated that conventional breeding efforts can be accelerated by using markers to locate genes controlling important agronomic traits and providing markers tightly linked to such genes for marker-assisted selection (MAS) at the seedling stage (see Figure 1). Marker-assisted selection is based on the principle that if a gene conferring a trait of interest is linked to an easily identifiable molecular marker, it may be more efficient to select for the marker than for the trait itself (Mazur and Tingey, 1995).
Figure 1. The Principle of Marker-Assisted Selection
Marker-Assisted Selection Conventional Breeding A 2 1
34 1 Week
B C 1 Year DNA marker known to be linked to rust resistance 1 2 3 4 134
2 Identification of Rust Resistant Individual
2 2 Production of New Varieties
A - Seedlings planted B - Presence of marker known to be near trait on map is detected in the laboratory, identifying disease resistant plant at seedling stage. C - After 1 year in the field, plant number 2 is shown to be disease resistant.
Two main strategies are commonly used to identify such markers for use in marker- assisted selection – genome mapping and Bulked Segregant Analysis.
Genome mapping strategies involve the use of molecular markers to construct genetic linkage maps that, in turn, provide a framework for the subsequent identification of genomic regions controlling important agronomic traits. Identification of these loci can be achieved by subjecting combined genotype and phenotype data to quantitative trait loci (QTL) analyses. Markers shown to display tight linkage with such trait loci can be then used for marker-assisted selection and for monitoring trait introgression.
26 In poplar, three types of molecular markers (Random Amplified Polymorphic DNAs (RAPDs), Restriction Fragment Length Polymorphism (RFLPs) and Sequence Tagged Sites (STSs)) have been used to construct maps of P. nigra and P. deltoides (Bradshaw et al., 1994). These markers have also been used to map monogenic resistance to Melampsora larici-populina and to roughly map a QTL for resistance towards Xanthamonas populi (Cervera et al., 1996a; 1996b; 1997). Whilst mapping efforts are underway in willow, no linkage maps have yet been published for any Salix species.
Although genetic mapping methods are extremely powerful for marker identification, construction of genetic maps is both labour-intensive and time-consuming, as production of large numbers of markers, spanning the entire genome at even intervals, is required. Consequently, short-cut approaches, such as Bulked Segregant Analysis, provide exciting opportunities. This technique for identification of markers linked to traits of interest, first described by Michelmore et al., (1991), does not require large numbers of previously mapped molecular markers. In brief, markers linked to traits are identified by comparative Amplified Fragment Length Polymorphism (AFLP) analysis of pooled DNA samples (bulks) from individuals showing extreme phenotypes, e.g. DNA from the most disease-resistant individuals is pooled, as is that of the most disease-susceptible individuals, and the AFLP profiles of both pools are compared. If a marker is present in the profile of one pool, but not the other, then the marker could potentially be linked to the phenotype. Confirmation of any potentially interesting markers is achieved by screening individuals within the bulks to see if the relationship between marker and phenotype still holds.
Types of Molecular Marker
Microsatellite Markers
Microsatellites or Simple Sequence Repeats (SSRs) are highly mutable loci that are abundant throughout the genome (Morgante and Olivieri, 1993) and consist of short sequence, tandemly repeated mono-, di-, tri-, tetra-, or penta-nucleotide repeating units. The hypervariablity of these units is thought to arise from strand slippage during DNA replication, leading to addition or deletion of repeat units. This, in turn, leads to a high degree of polymorphism at microsatellite loci, which can be characterised by PCR followed by electrophoresis to determine product length. Since, in most cases, the flanking regions of each microsatellite locus are unique, primers can be designed to these regions for use in amplification of particular loci (Rongwen et al., 1995).
Apart from their highly polymorphic nature, microsatellite markers possess numerous other favourable characteristics. They present a co-dominant marker system and thus can be used to detect heterozygosity, an advantage when working with an outcrossing species such as willow (Karp et al., 1996). They are also found to be abundant and show extensive coverage of the genome (Brown et al., 1996; Liu et al., 1995). The fact that polymorphisms at microsatellite loci are detected by PCR is also advantageous as relatively small amounts of starting material are required when compared to marker systems such as RFLPs. Also, microsatellite markers are robust
27 and can easily be exchanged between laboratories which is not the case with other types of molecular markers, such as RAPDs.
However, while microsatellite markers are extremely informative for each locus, and have many advantages associated with them, they are costly to develop both in terms of labour and resources. In general, microsatellite markers are non-transferable across genera, i.e. primer pairs developed in a particular genus may not amplify corresponding loci in another. Also, for applications such as genome mapping, where large numbers of markers are required, they do not produce as many data points per assay as dominant, multilocus marker systems (e.g. AFLPs, RAPDs).
AFLP Markers
The AFLP fingerprinting technique was first described by Zabeau and Vos (1993). Generation of AFLP markers involves digestion of genomic DNA with two restriction enzymes (a rare and a frequent cutting enzyme). Adapters are ligated to ends of these fragments, such that the adapter designed to the rare cutting site is biotinylated. Biotinylated fragments are then isolated using streptavidin beads and amplified by PCR using 33P–labelled primers designed to the frequent cutting adapter sequences. These primers allow selective amplification of a subset of the fragments as they extend, by 1 – 3 bp, into the restriction site and, therefore, only amplify those fragments in which the primer extensions match the nucleotides present in the flanking region of the restriction site. PCR products are visualised by polyacrylamide gel electrophoresis and autoradiography. In a later modification of the technique (Vos et al., 1995), the biotinylated step is removed and two successive rounds of PCR are carried out using primers with first one (in the first round) and then (in the second round) one to three selective nucleotides.
Although more technically complicated than some other marker systems, AFLPs provide a large number of data points per experiment making them especially useful for projects where large numbers of markers are required, such as genome mapping and genetic diversity studies (Barker et al., 1999). A further advantage associated with AFLP technology is the high level of reproducibility provided by a specific PCR- based system. However, the dominant nature of AFLP markers can be regarded as a disadvantage for mapping studies, as information about heterozygosity is not provided.
Independently, microsatellites and AFLPs have limitations as markers for mapping but when used in combination they complement one another to meet the requirements of this project.
PROJECT OBJECTIVES
The main objectives of this project are as follows:
• Construct a highly saturated genetic linkage map of willow population K3
28 • Locate genomic regions controlling important agronomic traits
• Provide markers linked to traits for marker-assisted selection
Traits of Interest
Important agronomic traits to be examined include resistance to Melampsora rust, resistance to Chrysomelid beetle herbivory and yield. The genetic basis of resistance to pest and pathogen pressures is poorly understood at present, and may be quantitative (under the control of a large number of genes) or qualitative (controlled by a small number of genes each with large effect). It is assumed that yield is a quantitative trait.
CONSTRUCTION OF THE K3 GENETIC MAP
The K3 mapping population, comprising 66 individuals, was produced by Svalöf Weibull AB from a full-sib cross of Salix viminalis. In a previous project at Long Ashton Research Station (LARS), funded under AIR (AIR2-CT92-1617), Dr. J.H.A. Barker initiated construction of a first AFLP/SSR map of Salix using this population. In this project, additional marker information has been added to the original data set and a preliminary linkage map constructed. Development and mapping of further markers has continued to ensure all map regions are sufficiently saturated. This will increase the likelihood that, when identified, genomic regions controlling traits of interest will be positioned sufficiently close to a marker for marker-assisted selection to be viable.
While this population is suitable for genetic linkage mapping, the relatively small population size means that it does not lend itself to trait mapping where large progeny numbers are favoured. Furthermore, for mapping, genes of interest must be segregating in the mapping population. It is unknown whether this is case in K3. To circumvent these problems, two additional mapping populations, K1 and K8, have been established.
MAPPING POPULATIONS K1 AND K8
From 18 potential crosses growing as part of the European Willow Breeding Programme, K1 and K8 were selected on the basis of seed availability and observed parental segregation for rust and beetle resistance in the field. The pedigrees of these crosses are shown in Figure 2.
Figure 2. Pedigrees of K1 and K8 Crosses (R=resistant to rusts (field assessed) and S=susceptible).
K1: (Cross 591)
29 Great grandparents: Astrid (SW880435) x Bjorn (SW910006) (S. viminalis) (S. viminalis x S. schwerinii) | Grandparents: Astrid (SW880435) x SW 930984 (S. viminalis) (S. viminalis x S. schwerinii)
Parents (F1): R6 x R11 | Progeny (F2): 946 planted in trial
K8: (Cross 598)
Great grandparents: Astrid (SW880435) x Bjorn (SW910006) (S. viminalis) (S. viminalis x S. schwerinii) | Grandparents: Astrid (SW880435) x SW 930984 (S. viminalis) (S. viminalis x S. schwerinii)
Parents (F1): S1 x R11 | PROGENY (F2): 947 PLANTED IN TRIAL
The large progeny sizes of these populations make them more amenable to trait mapping. Initially, two families were established for this purpose as there was some risk in ensuring that good segregation would be achieved for all of the traits to be examined. Following initial assessments one family will be selected for detailed analyses.
Phenotypic Assessments
In order to identify regions of the genome controlling a particular trait, phenotypic data is required. This information can be used to identify individuals with extreme phenotypes for use in bulked segregant analysis or, alternatively, can be combined with genotypic data for QTL analysis. Field assessments for resistance to rust, resistance to beetle feeding will be carried out and a number of yield measurements taken, including number of shoots per stool, stem diameter, stem height and fresh and dry weights.
In addition to the planned field-based assessments, laboratory-based phenotypic assays have been performed as mixed populations of rust pathotypes and beetle
30 species may be present in the field at time of assessment, making interpretation of field assessment data complex.
Rust resistance/pathogen virulence interactions, initially for two prevalent UK pathotypes, LET1 and LET5, were characterised for a random sample of 100 individuals from both K1 and K8 crosses, using a leaf disc inoculation technique previously developed at LARS (Pei et al., 1996). To determine the level of resistance presented by each individual, the number and size of pustules present and the number of spores produced are being recorded. Phenotypic data will be used for Bulked Segregant Analysis and QTL analyses.
Beetle feeding assays were performed according to a procedure previously developed at LARS. In brief, willow beetles, of a single species, were allowed to feed on a collection of leaf discs prepared from individual progeny of the population under examination. Feeding preferences were then determined by analysis of leaf disc area remaining following the feeding period, a measure of the level of susceptibility/resistance displayed by each population. Again, phenotypic data from these assays will be used for Bulked Segregant Analysis and QTL analyses.
Preliminary data from these experiments suggest that both K1 and K8 populations are segregating for rust and beetle resistance.
Overall Marker Identification Strategy
The marker identification strategy used in this project is summarised in Figure 3. In brief, the K3 mapping population has been used to construct a highly saturated linkage map using AFLP and microsatellite markers. A K1 or K8 framework linkage map will be constructed and anchored to the K3 linkage map using microsatellite markers common to both populations. Using QTL analysis, the K1 or K8 framework map will be used to map traits of interest, with the K3 map providing more detailed marker position data required for identification of closely linked markers. Furthermore, in an attempt to merge the mapping efforts of willow with poplar, the willow maps will be anchored to an existing poplar linkage map, again using microsatellite markers. It is thought that willow and poplar genomes possess a high level of synteny (the order of genes may be similar), and as markers linked to rust resistance and a variety of yield parameters have already been mapped in poplar, this existing information may assist in the identification of corresponding markers in willow. An integrated map will result, incorporating dense marker information from K3, mapped traits from K1/K8 and existing marker and trait data from poplar. Bulked segregant analysis will also be used as a potential shortcut for marker identification.
31 Figure 3. Overall Marker Identification Strategy.
Cross K3 mapping Poplar K1 or K8 BSA population mapping
Poplar Map
Anchor K3 and Trait maps Anchor K3 and Poplar maps using SSRs using SSRs Trait Map K3 Map SSRs AFLPs & SSRs
Identification Integrated map of Markers for AFLPs, SSRs, Traits MAS
Identifying Leaf and Stem Traits Relevant to Biomass Yield in Willow
Currently, there is a lack of information regarding which physiological leaf and stem traits may be indicative of yield for willow. As part of this project, this question is being addressed by Kathryn Robinson and Rachel Ferris of Gail Taylor’s Group at the
32 University of Southampton. This group has a high level of expertise in this area having undertaken similar, detailed studies in poplar.
To determine whether leaf and stem traits that correlate with high biomass yield in poplar are applicable to willow, physiological comparisons of these traits are being performed on consistently high and low yielding willow genotypes from the Comparative Yield Trials at LARS. Traits under examination include stem height, stem diameter, number of shoots per stool, leaf extension rate, leaf production rate, leaf area and a number of photosynthetic characteristics. Initial results suggest that leaf extension rate, a good indicator of yield in poplar, may be applicable to willow.
Once identified, informative traits for yield will be assessed in the K1 or K8 mapping population and the data used in QTL analysis to identify markers linked to these traits for use in future selections.
The Future
Marker production will continue in order to saturate the K3 map and to permit construction of K1 or K8 framework maps. Further phenotypic assessments, both laboratory- and field-based, will be carried out to obtain data for use in marker identification by QTL analysis and Bulked Segregant Analysis. Any markers potentially linked to traits of interest will then be further tested in the laboratory, and ultimately validated for use in simple assays for marker-assisted selection by the willow breeder.
Acknowledgements
This project is supported by the DTI’s New & Renewable Energy Programme (managed by ETSU) [Agreement No. B/W6/00599/00/00]. The authors thank Dr Kevin Lindegaard (European Willow Breeding Programme) for the crosses and Drs. Ming Pei, Tom Hunter and Lori Peacock at Long Ashton Research Station for help with the rust and beetle assays. IACR receives grant-aided support from the Biotechnology and Biological Sciences Research Council of the United Kingdom.
References
33 Barker, J.H.A., Matthes, M., Arnold, G.M., Edwards, K.J., Åhman, I., Larsson, S. and Karp, A. (1999). Characterisation of genetic diversity in potential biomass willows (Salix spp.) by RAPD and AFLP analyses. Genome 42, 173-183.
Bradshaw, H.D., Villar, M. and Watson, B.D. (1994). Molecular genetics of growth and development in Populus. 3. A genetic linkage map of a hybrid poplar composed of RFLP, STS and RAPD markers. Theoretical and Applied Genetics 98, 167-178.
Brown, S.M., Hopkins, M.S., Mitchell, S.E., Senior, M.L., Wang, T.Y., Duncan, R.R., Gonzalez-Candelas, F. and Kresovich, S. (1996). Multiple methods for the identification of polymorphic simple sequence repeats (SSRs) in sorghum [Sorghum bicolor (L.) Moench]. Theoretical and Applied Genetics 93, 190-198.
Cervera, M.T., Storme, V., Liu, B., Gusmao, J., Steenackers, M., Ivens, B., Michiels, B., Van Montagu, M. and Boerjan, W. (1997). AFLP™ genome mapping of poplar. Med. Fac. Landbouww. Univ. Gent 62/4a, 1435-1441.
Cervera, M.T., Gusmao, J., Steenakers, M., Van Gysel, A., Van Montagu, M. and Boerjan, W.T. (1996a). Application of AFLP-based molecular markers to breeding of Populus spp. Plant Growth Regululation, 20, 47-52.
Cervera, M.T., Gusmao, J., Steenackers, M., Peleman, J., Storme, V., Vanden Broeck, A., Van Montagu, M. and Boerjan, W. (1996b). Identification of AFLP markers for resistance against Melampsora larici-populina in Populus. Theoretical and Applied Genetics, 93, 733-737.
Karp, A., Seberg, O. and Buiatti, M. (1996). Molecular techniques in the assessment of botanical diversity. Annals of Botany 78, 143-149.
Liu, Z-W, Jarret, R.L., Kresovich, S. and Duncan, R.R. (1995). Characterization and analysis of simple sequence repeat (SSR) loci in seashore paspalum (Paspalum vaginatum Swartz). Theoretical and Applied Genetics 91, 47-52.
Mazur, B.J and Tingey, S.V. (1995). Genetic mapping and introgression of genes of agronomic importance. Current Opinion in Biotechnology 6, 175-182.
Michelmore, R.W., Paran., I. and Kesseli, R.V. (1991). Identification of markers linked to disease-resistance genes by bulked segregant analysis: A rapid method to
34 detect markers in specific genomic regions by using segregating populations. Proceedings of the National Academy of Sciences USA 88, 9928-9832.
Morgante, M. and Olivieri, A.M. (1993). PCR-amplified microsatellites as markers in plant genetics. The Plant Journal 3, (1),175-182.
Pei, M.H., Royle, D.J. and Hunter, T. (1996). Pathogenic specialization in Melampsora epitea var epitea on Salix. Plant Pathology 45, (4), 679-690.
Rongwen, J., Akkaya, M.S., Bhagwat. A.A., Lavi, U. and Cregan, P.B. (1995). The use of microsatellite DNA markers for soybean genotype identification. Theoretical and Applied Genetics 90, 43-48.
Vos, P. Hogers, R., Bleeker, M., Rijans, M., Van de Lee, T., Hornes, M., Frijters, A., Pot, J., Peleman, J., Kuiper, M. and Zabeau, M. (1995) AFLP: a new technique for DNA fingerprinting. Nucleic Acids Res 23:4407-4414
Zabeau, M. and Vos, P. (1993). Selective restriction fragment amplification : a general method for DNA fingerprinting. European Patent Application, publication no : EP 0534858-A1, No. 92402629.7.
35 The use of willow mixtures as a disease control strategy
A. R. McCracken1,3, A. C. Bell1,3, W. M. Dawson2,3 & R. J. Marks1,3 E-mail: [email protected] 1Applied Plant Science Division, Dept. of Agriculture and Rural Development Northern Ireland, Newforge Lane, Belfast BT9 5PX, N. Ireland UK 2Northern Ireland Horticulture and Plant Breeding Station, Dept. of Agriculture and Rural Development, Loughgall, Co. Armagh, BT61 8JB, N. Ireland, UK 3Dept. of Applied Plant Science, School of Agriculture and Food Science, Queen’s University Belfast, Newforge Lane, Belfast BT9 5PX, N. Ireland, UK
INTRODUCTION
Rust, caused by Melampsora epitea var. epitea, is probably the greatest limitation to the growth of willow (Salix spp.) in short rotation coppice for biomass. The disease can affect a wide range of Salix species and varieties. In the most susceptible varieties rust causes heavy premature leaf fall followed by infection by a whole spectrum of secondary pathogens which can result in death of shoots and even stools (McCracken & Dawson 1992). In N. Ireland rust was first observed on S. burjatica ‘Korso’ in 1986. Within a few years this variety was so badly affected that often 60% of stools were killed by the time of flushing the spring following harvest (Dawson & McCracken, 1993). While there is a range of susceptibilities to rust a number of varieties have become so susceptible that they can no longer be grown commercially in many parts of the country.
Short rotation coppice willow is a long-term crop and therefore is it essential to have a disease control strategy which reduces the impact of the rust and at the same time ensures the sustainability of plantations over the life of the crop, which may be up to 25 years. Although rust can be controlled by the intensive use of fungicides (McCracken & Dawson, 1990; 1997) it was clear that this was not a viable approach to disease control. Early work reported by Wolfe (1985) indicated that mixtures of multilines or cultivars of cereals was an effective method of reducing the impact of diseases, especially those caused by obligate pathogens such as rusts and mildews. Therefore in the mid 1980s trials were planted in N. Ireland which comprised mixtures of willow species and varieties, in order to test their effectiveness in reducing rust. These early trials indicated that when a susceptible willow variety was included in a mixture the first appearance of rust was delayed, the build up of the disease slowed and the final levels of rust significantly reduced compared to the same variety growing in a mono-culture (McCracken & Dawson 1997). It was also reported that dry matter yields from mixtures were consistently higher than the mean of the constituent varieties grown in mono-culture (McCracken & Dawson 1994). The subsequent trials, reported in this paper, were planted with a view to determining the effect of numbers of varieties within a mixture, the impact of planting density and the performance of some of the newer planting material within mixtures.
In most seasons many of the willow varieties planted carried a heavy infestation of willow beetles (Sage & Tucker, 1998) which could cause significant feeding damage. Observations and measurements were carried out to assess the level and importance of this damage.
TRIALS
Plantations
36 The plantations were established at Castlearchdale, Co. Fermanagh, N. Ireland (54028’N 7043’W), 60-70m above sea level and on a site with heavily gleyed soils with impeded drainage. The trial was planted in two replicate blocks. Block 1 was planed in spring 1994 and block 2 in spring 1995. Both blocks were cut back at the end of their establishment year and block 1 again in 1995/96 meaning that in the summer 1996 both blocks were at the same stage, i.e. regrowth from freshly coppice stools. Twenty Salix spp varieties (Table 1) were planted in large mono-culture (32 x 14m in Block 1 and 32 x 11m in Block 2) in double rows in a split plot design with three densities - 10,000ha-1, 15,000ha-1 and 20,000ha-1. Four mixture combinations were employed comprising 5, 10, 15 and 20 varieties (Table 1).
Table 1: List of Salix spp. varieties planted in 5, 10, 15 or 20 way mixtures
Variety Variety Mixture No. 3 S. burjatica Germany 5 10 15 20 4 S. mollissima-undulata SQ83 5 101520 13 S. dasyclados x aquatica V7511 5 10 15 20 17 S. viminalis 77082 5 10 15 20 26 S. dasyclados x caprea V794 5 101520
14 S. viminalis x aquatica V7503 - 101520 23 S. viminalis 78118 - 10 15 20 28 S. viminalis 78183 - 10 15 20 32 S. schwerinii x viminalis x dasyclados V7531 - 10 15 20 48 S. viminalis 870146 ULV - 10 15 20
12 S. viminalis x caprea V789 - - 15 20 18 S. viminalis 77683 - - 15 20 21 S. viminalis 78101 - - 15 20 29 S. viminalis 78195 - - 15 20 34 S. schwerinii x aquatica V7534 - - 15 20
1 S. viminalis 77699 - - - 20 6 S. viminalis Gigantea - - - 20 9 S. viminalis Gustav - - - 20 45 S. schwerinii x aquatica V7533 - - - 20 46 S. viminalis 870082 ORM - - - 20
All mixtures were planted as fully random intimate mixtures. At the time of planting a map was prepared of the planting pattern so that subsequently each stool could be identified.
Disease assessments Starting in mid June and at two week intervals leaf samples were collected. At least 100 leaves were taken from approximately ten stools of each variety growing in mono-culture and in mixtures. The percentage number of leaves with rust was determined and the percentage cover of each leaf assessed using a disease assessment key (McCracken & Dawson 1992). These assessments were carried out each year from 1996 (first year growth from freshly coppiced stools), 1997 (second year growth), 1998 (third year growth), 1999 (first year growth from freshly coppiced stools) stools and 2000 (second year growth). By using the number of infected leaves and the percentage rust cover, a cumulative rust score was calculated (McCracken & Dawson 1992).
37 Pest assessments In 1999 and 2000 the same leaf samples which had been scored for rust were assessed visually for willow beetle damage. Herbivory was scored as the percentage of each leaf removed and a mean score was calculated for each variety in mono-culture and in mixtures.
RESULTS & DISCUSSION
Disease In general the build up of disease was slower in mixtures than it was in the mono-varietal plots. The different Salix spp. varieties had a range of susceptibilities to rust, which in turn influenced the effect exerted by inclusion within a mixture. Planting density had no discernible impact on the incidence or severity of disease. The effect of numbers of components in a mixture on rust development on any individual Salix variety within a mixture tended not to be consistent. However there are other benefits in having larger numbers present within a mixture. Pathotype diversity on an individual variety was normally greater when that variety was in a mixture (McCracken et al. 2000). This increased diversity (Fig 1) in turn increased competition between pathotypes, which appeared to reduce selection pressures, which may have resulted in the development of super races of Melampsora epitea var. epitea. Furthermore, yield compensation for stools that had died within a mixture, due to disease or other causes, was more effective in the mixtures with a larger number of components.
LSD = 0.068 0.6
D i 0.4 v e r s i t 0.2 y
0 Mono Mix
Fig 1: Melampsora epitea pathotype diversity (Shannon-Weaver Index) on Salix spp. varieties grown as mixtures and as mono-culture (McCracken et al. 2000).
The pattern of rust development on individual Salix spp. varieties when included within a mixture was largely determined by their level of susceptibility to rust.
Salix burjatica Germany Salix dasyclados x aquatica V7511 Salix dasyclados x caprea V794
S. burjatica Germany has been grown in N. Ireland for almost twenty years. Rust was first recorded on this variety at the end of the growing season on 1987. In subsequent seasons rust was observed, usually towards the middle or end of the summer, on the upper leaves and at low levels, inflicting little or no damage to plants (McCracken & Dawson 1992). However in 1996 S. burjatica Germany was severely infected with rust, to a degree that, following premature leaf fall there was extensive rod and stool death in the mono-varietal plots. In the period from 1996 – 2000 this variety has been so badly affected that it has been almost totally killed in some of the mono-varietal plots. Its behaviour in mixtures has been erratic. The build up of rust has been slowed down on plants growing in the larger (10 – 20
38 varietal mixtures) although still causing significant damage and loss of stools. While S. dasyclados x aquatica V7511and S. dasyclados x caprea V794 have not been grown in N. Ireland for as long as S. burjatica Germany the pattern of their demise has been similar.
It has been concluded that where a variety is, or becomes particularly susceptible to rust, especially in conditions of high disease pressures, its inclusion in a mixture may slow down the effects of the disease but is probably not enough to maintain it as a productive variety.
POLY X5 POLY X10 POLY X15 E
10000 POLY X20 MONO D1
5000 MEAN ACCUMULATIVE RUST SCOR RUST ACCUMULATIVE MEAN
0 08/06/99 22/06/99 07/07/99 21/07/99 03/08/99 17/08/99 31/08/99 14/09/99 DATE
Fig 2: Melampsora epitea rust disease development in 1999 on S. dasyclados x aquatica V7511 grown in mono-culture and included in 5, 10, 15 and 20-way mixtures.
S. mollissima-undulata ‘SQ83’
In 1996 S. mollissima-undulata SQ83 was severely affected by rust, starting early in the growing season and followed by significant death of rods when in mono-culture. In that year inclusion within all mixtures had a significant effect in reducing disease. In subsequent years the disease followed a different pattern. It tended not to become evident until much later, often late August and consequently its effect on growth was less. There was a rapid increase in rust scores in mono-culture which were significantly less in mixtures.
S. viminalis ‘77082’, ‘78118’, ‘78183’, ‘870146 ULV’, ‘77683’, ‘78101’, ‘78195’, ‘Gigantea’, ‘Gustav’, and ‘870082 ORM’.
In each year from 1996 – 1999 S. viminalis ‘77082’ carried low levels of disease, especially early in the growing season, up until September. Only in 1996 were significant levels of rust observed in July. In none of the years were any differences recorded in the pattern of rust development in the mixtures or on plants grown in mono-culture.
In all years S. viminalis ‘78118’ had only low levels of rust, normally only evident as the season progressed. Inclusion in mixtures appeared not to affect the pattern of disease development.
S. viminalis ‘78183’ had low levels of disease in all seasons with a general trend of there being more rust on plants grown in mono-culture.
S. viminalis ‘870146 ULV’ had virtually no rust in any year with the exception of 1996, when levels of disease on mixtures was less than in mono-culture.
39 S. viminalis ‘77683’. In all years only infrequent pustules were observed at the end of the growing season in late September. This may have been partly due to this variety being very variable and almost constituting a mixture in its own right.
S. viminalis ‘78101’ tended to have a few pustules of rust, even in early July. However there was only a slow increase in the amount of disease as the season progressed and at no point was there a difference between the amount of disease on this variety whether grown in mono-culture or within a mixture.
S. viminalis ‘78195’ had low levels of disease which only became apparent towards the end of August. No significant differences were observed between disease levels in mixtures or mono-culture.
S. viminalis ‘77669’ had low levels of disease in all years.
S. viminalis ‘Gigantea’ had no significant levels of rust observed in any year.
S. viminalis ‘Gustav’ had low to moderate rust in each year. In 1997 it was observed early while in 1998 and 1999 rust pustules were not recorded till much later. The general trend at each recording date was for reduced disease levels in mixtures compared to mono-culture.
S. viminalis ‘870082 ORM’ had only very low levels of rust in any year.
Salix viminalis mixtures Each of the straight S. viminalis varieties grown in the trial had only low susceptibility to rust, and often were only affected late in the season. There were however different patterns of disease development. It can therefore be argued, that even a straight S. viminalis mixture in which its components may be susceptible to the same Melampsora epitea pathotypes will still act as a genuine mixture in terms of reducing disease and reducing the selection pressures on the pathogen. This is the topic of on-going research. A S. viminalis trial has been established at the Northern Ireland Horticulture and Plant Breeding Station, Loughgall in order to evaluate disease development within a mixed plantation where all of the components are straight S. viminalis selections.
S. schwerinii x viminalis x dasyclados V7531 S. schwerinii x aquatica V7533 S. schwerinii x aquatica V7534
Varieties with S. schwerinii parentage tend to display a high level of resistance to rust. Since planting the trial in 1996 rust has only been observed at minimal levels (occasional pustules on < 0.1% of leaves) in any year, irrespective of age of growth.
There was obviously no immediate disease reduction advantage in including such varieties within a mixture. It can be argued however that growing them in mixtures reduces the potential disease pressures and hence ensures their long-term sustainability over the full 25 – 30 year life of a plantation.
Yield
The mean yield from 100m-2 of any mixture was significantly greater than the mean yield from an equivalent area of the constituent varieties grown in mono-culture (Fig. 3). With the exception of the five-way mixture, which was significantly lower, there were no significant yield difference between the 10, 15 and 20 way mixtures. These trends were consistently observed at each of the three densities.
Five-way mixture The yield from the five-way mixture was low because of the death or weak growth of three of the components – S. burjatica Germany, S. dasyclados x aquatica V7511, S. dasyclados x caprea V794, due to the impact of the continuous rust infections. The two remaining varieties had made major compensation for the extra space. Variety 17, S. viminalis 77082 produced almost 50% more dry matter in this mixture compared to its growth in mono-culture. Similarly Variety 4, gave a 20% increase in yield.
40 250 Mono Mi xture
200
Y 150 i e l d 100
50
0 X 5 X 10 X 15 X 20 Mixture
Fig. 3: Dry matter yield (kg) from fixed area (100m2) of mixtures of Salix spp. varieties compared to an equivalent area of the components grown as mono-culture.
Ten-way mixture Yield from the ten-way mixture was significantly higher than the equivalent from the constituent varieties. This was despite the demise of S. burjatica Germany, S. dasyclados x aquatica V7511, S. dasyclados x caprea V794, whose growth was significantly affected by rust. S. viminalis 77082, S. schwerinii x viminalis x dasyclados V7531 and S. viminalis 870146 ULV all contributed significantly more dry matter yield than had been predicted.
Fifteen-way mixture The increased yield was due mainly to the increased contribution from S. schwerinii x viminalis x dasyclados V7531, S. viminalis 870146 ULV, S. viminalis 77683, S. viminalis 78195 and S. schwerinii x aquatica V7534. S. viminalis x aquatica V7503 and S. viminalis x caprea V789 both contributed significantly less in this mixture than might have been predicted. Susceptibility to rust may have been a contributory factor. However their poor performance in mixtures was due more to their lack of ability to compete with other components. It has been observed that in the second growth cycle S. viminalis x caprea V789 has had a low survival rate in the 15 and 20-way mixtures.
Twenty-way mixture The major contributors to the yield from this mixture were S. viminalis 77683, S. viminalis 78195, S. schwerinii x aquatica V7534, S. schwerinii x aquatica V7533 and S. viminalis 870082 ORM.
Willow beetle damage The levels of damage to leaves from the various varieties in mono-culture are shown in Fig.4. Overall, damage levels were moderate. However, five varieties exhibited damage, which was significantly greater than the others. These were S.viminalis 78118, S.viminalis 870146 ULV, S.viminalis x caprea V789, S.viminalis 77683 and S.viminalis Gigantea. Preferential feeding of Phratora vulgatissima on different varieties of willow is documented in the literature (Sage and Tucker, 1998; Kearns, 1934), and
41 the results in the present trial indicate that these five varieties could be at risk of significant damage in years when willow beetle populations are high.
When the results from the mixture plots were examined, these varieties showed a trend towards reduced damage with increasing numbers of other varieties in the mixtures. Although this result did not achieve statistical significance, it does suggest that the inclusion of damage-susceptible varieties in mixed plots may offer them some protection from willow beetle attack. It is likely that this would have become more obvious if beetle populations had been higher during the period of this study.
10 9 8 7 6 5 4 % leaf damage 3 2 1 0 3 4 13 17 26 14 23 28 32 48 12 18 21 29 34 1 6 9 45 46
Variety Number Fig. 4: Willow beetle damage to varieties in mono-culture. The dark bars indicate those varieties that show significantly higher levels of damage.
CONCLUSIONS AND RECOMMENDATIONS
It is clear that there was significant yield benefits from growing willow in inter- and intra-species mixtures. This is partially due to reduced disease levels on some varieties when grown in a mixture. It is also due to physiological interactions, which encourages some varieties to grow more vigorously. Furthermore if a variety dies out of a mixture either due to disease, competition or some other factor then the remaining varieties can compensate for their loss. For this reason larger mixtures are more beneficial. In this current trial 60% of the five-way mixture was killed and while the two remain varieties did make major compensations it was not sufficient. However in the twenty-way mixture where only 15% was lost the impact on yield was much less obvious. The authors also consider than random intimate mixtures give much greater opportunity for compensation compared to either line mixtures or mosaics. The use of short-run mixtures is a compromise to the needs of commercial planting.
When planting willow plantations it is therefore recommended:
42 • Plant inter- and intra-species mixtures of willow. • Plant as diverse a mixture as possible. • Plant mixtures containing least 6 – 10, if possible. • Avoid varieties which are rust susceptible or have become rust susceptible in other locations. • Mixtures should be as random as commercial practice permits.
REFERENCES
Dawson, W.M. & McCracken, A.R. (1993). The effect of Melampsora rust on the growth and development of Salix burjatica ‘Korso’ in Northern Ireland. European Journal of Forest Pathology 24, 32 – 39.
Kearns, H.G.H. (1934). The control of insect pest of basket willows. Annual Report of Long Ashton Research Station, 1934, 3-5.
McCracken, A.R. & Dawson, W.M. (1992). Clonal response in Salix to Melampsora rusts in short rotation coppice plantations. European Journal of Forest Pathology 22, 19 – 28.
McCracken, A.R. & Dawson, W.M. (1990). The effect of Melampsora sp rust on growth an development of Salix in Northern Ireland. Proceedings of Joint Meeting IEA Task 5 Activity Groups on the exchange of Genetic Material and Joint Trials on Alnus, Populus and Salix and Pests / Disease Management, Poplar Research Centre, Geraardsbergen, Belgium 11 – 14 September 1990, Ed. By V. Steenackers. Geraardsbergen: International Energy Agency. pp 15 – 26.
McCracken, A.R. & Dawson, W.M. (1994). Experiences in the use of mixed-clonal stands of Salix as a method of reducing the impact of rust disease. Norwegian Journal of Agricultural Science 18 (suppl.) 101 – 109
McCracken, A.R. & Dawson, W.M. (1997). Growing clonal mixtures of willow to reduce the effect of Melampsora epitea var. epitea. Europena Journal of Forest Pathology 27, 319- 329.
McCracken, A. R., Dawson, W. M., Watson, Sally &. Allen, Chanel Y (2000). Pathotype composition in Melampsora epitea populations occurring on willow (Salix) grown in mixed and mono-culture plantations. European Journal of Plant Pathology. (In press).
Sage, R.B. & Tucker, K. (1998). The distribution of Phratora vulgatissima (Coleoptera: Chrysomelidae) on cultivated willows in Britain and Ireland. European Journal of Forest Pathology, 28, 289-296.
Wolfe, M. S. (1985). The current status and prospects of multiline, cultivars and variety mixtures for disease resistance. Annual Review of Phytopathology 23, 251 – 273.
43 Breeding Strategies – The Way Forward
KEVIN LINDEGAARD
IACR-Long Ashton Research Station, Department of Agricultural Sciences, University of Bristol, Long Ashton, Bristol BS41 9AF, UK E-mail: [email protected]
Abstract
The European Willow Breeding Programme was initiated in 1996 to produce new superior willow varieties for biomass energy. To date, the crossing programme has focussed on inter-species combinations between a base population of approximately 100 varieties obtained from the UK National Willows Collection held at IACR-Long Ashton. In total, 750 crosses have been made and over 40,000 unique genotypes have been screened in a seedling nursery. Selected individuals have been evaluated in an extensive array of breeding trials and superior genotypes have been advanced to yield trials. Currently, over 50 genotypes are in advanced observation and yield trials. Of these it is expected that a small proportion will be multiplied for commercial use in the biomass industry from 2004 onwards. The current methods used to select willows in the field are unsatisfactory and need urgent refinement. Future selection strategies should attempt to combine the breeders’ observations with quantitative data and yield estimates. Such information will enable multiple trait selection strategies to be employed. The construction of a genetic map of willows and the identification of markers linked to traits of interest are extremely important developments. The use of molecular marker technology will facilitate the rapid selection of desirable genotypes at the seedling stage and accelerate the introgression of valuable traits in backcross procedures. In future hybridisation schemes, short term breeding initiatives should be coupled with longer-term recurrent selection strategies. Other long-term prospects might include the production of inbred lines by anther culture, polyploid breeding and the application of GM technology.
Key words: Salix, willow, Short Rotation Coppice, biomass, breeding, interspecific cross, recurrent selection.
INTRODUCTION
The European Willow Breeding Programme (EWBP) was established in 1996 to produce new willow varieties principally for biomass energy production (Lindegaard and Barker, 1997). As with all plant breeding programmes success in willow breeding depends on the creation of variability through hybridisation combined with effective methods of assessment and selection of the progeny in the field. During the first five years of the programme various breeding and selection strategies have been employed. The purpose of this paper is to review the relative success of these strategies and consider where procedures may be improved and whether new approaches are necessary to accelerate the production of varieties yet further.
44 Crossing Programme and Field Evaluations 1996-2000
Between 1996-2000, over 750 crosses have been made using a base population of 100 individuals belonging to 20 distinct willow species from North and Central Europe, Russia, Asia and North America which are maintained in the UK National Willows Collection held at IACR-Long Ashton. 40,000 unique willow genotypes have been screened in a seedling nursery and selected on the basis of yield estimations, resistance to disease and pests and erect growth habit. Between 5-10% of these selections have been advanced initially to field- based observation trials located at IACR-Long Ashton, from which elite breeding lines have been identified and subsequently included in smaller multi-site observation and yield trials. The breeding scheme being followed is demonstrated in Figure 1 and the progress of the breeding programme to date is illustrated in Figure 2. Already, over 50 genotypes are in advanced observation and yield trials. Of these it is expected that a small proportion will be multiplied for commercial use in the biomass industry from 2004 onwards. It is hoped that these will help increase the average yield to around 15 dry tonnes per hectare per year.
The Way Forward
In only a very short time giant strides have been made in the willow breeding programme. For instance, the success rate of the crosses performed has increased year on year and currently stands at 79.5%, almost four times that achieved in 1996. However, there are a number of shortcomings that have not received sufficient attention. These include field selection strategies and the development of medium- to long-term breeding initiatives. These areas will be considered in turn below along with key research requirements and the potential for novel applications of SRC willows.
Selection Strategies
Until now, field selections from observation trial 1 have been based solely on the value judgements made by the breeder. In future, if we are to be absolutely sure that all the best genotypes are selected for advancement we should not merely rely on “breeders’ eye” alone, but must support this expertise with quantitative data such as measurements of height, stem diameter, number of shoots etc. It is hoped that from 2000 onwards the use of one-hand computer callipers (Savcor Ltd) will provide a rapid non-destructive technique for estimating the yield of genotypes in observation trial 1. The technique used will be that performed on the national trial network by Forest Research (see Armstrong, this volume). The information gained from such an exercise will pave the way for more informed selection procedures for SRC willows.
As with most crops, willow breeding and selection schemes attempt to select for many traits simultaneously. Characters of interest include yield, disease and pest resistance, growth habit and wood quality. Trying to select genotypes that fulfil such
45 diverse criteria from observations alone is extremely demanding and subject to error. Therefore, in future, it would be sensible to consider the utilisation of a multiple trait selection strategy that aims to maximise genetic gain. There are a number of methods that endeavour to deal with this problem, such as index selection, independent culling, tandem selection and principal component analysis (Bisoffi and Gullberg, 1996) as well as cluster analysis (Ares and Gutierrez, 1996). These all require detailed measurements of several traits of interest. Cotterill and Dean (1990) suggest that index selection is the most efficient method for successful tree breeding. However, Bisoffi and Gullberg (1996) report no examples of index selection being used in poplar breeding. Instead the use of principal component analysis (Bisoffi, 1990) and cluster analysis (Ares and Gutierrez, 1996) have shown promise and should thus be considered most relevant to willow breeding. A distinct, albeit related concept involves the development of a tree ideotype (Dickmann, 1985). Dickmann and Keathley (1996) propose an ideotype selection programme in poplar. In this example, the breeder selects a subset of genetically correlated characters and these are combined as a single quantitative trait. This working ideotype is then treated as one of two to three variables in an independent culling process, in which clones that do not achieve a certain predetermined standard are eliminated. Although there are only a few examples of ideotype breeding in practice, Koski and Dickmann (1992) suggest encouraging results from the use of a modelling approach with SRC willows and poplars.
In the medium term, the construction of a genetic map for willow and the identification of markers linked to traits of interest will facilitate the rapid selection of desirable genotypes at the seedling stage and accelerate the introgression of valuable traits in backcross procedures. This technique, termed marker-assisted selection, will improve the efficiency of the initial stages of the willow breeding process (see Hanley et al, this volume).
Medium-Long Term Breeding Initiatives
Figure 3 illustrates the various hybridisation pathways so far utilised in the willow breeding programme. The vast majority of the crosses fall into only three categories: interspecific crosses between two species (type b), three-way species crosses (type f) and four-way species crosses, also known as double crosses (type g). Together these comprise over 90% of the crosses attempted. These types of crosses are beneficial because they maximise heterozygosity and the breeder may select specific gene combinations demonstrating heterosis. Hybrid vigour in such crosses is due to the trapping of dominant and epistatic gene action without eliminating or masking favourable recessive genes (Stettler, Zsuffa and Wu, 1996). However, in many cases this approach will result in barriers to continuing gains from genetic improvement because of a reduction in crossability caused by differences in ploidy levels and pollen-pistil incongruity in certain species (Mosseler, 1990). As a result of this, continuing solely with this approach would necessitate retracing successful crosses with progressively poorer parents due to the exhaustion of the germplasm available.
46 Hence, it is imperative that we back up such productive short-term approaches with longer-term strategies based on population improvement. Long-term breeding strategies aim to increase genetic diversity and create new crossing opportunities for the future (Kang, Lascoux and Gullberg, 1996). This can be effectively achieved using recurrent selection schemes which involve the concentration of desirable alleles during repeated cycles of hybridisation and selection (Jenkins, 1940; Hull, 1945). Several forms of recurrent selection exist but the application best suited for tree crops is termed reciprocal recurrent selection. Baudouin et al (1997) reviewed the breeding strategies of six tropical tree crops (cacao, coffee, rubber tree, oil palm, coconut and eucalyptus) and showed that all but rubber tree improvement were based on this method. Also this is the preferred strategy for long term poplar breeding (Stettler, Zsuffa and Wu, 1996; Bisoffi and Gullberg, 1996).
Figure 4 illustrates a generalised model for reciprocal recurrent selection for general hybridising ability (RRS-GHA) as used in the Italian poplar breeding programme (Bisoffi, 1990). This breeding system has the advantage of capturing both additive and non-additive gene effects together with the simultaneous improvement of two base populations and their hybrid offspring. The downside of this strategy is that it is labour intensive and demanding of field space.
Other long-term initiatives are also required for the continued success of the breeding programme. These include developing protocols for the efficient chromosome doubling of triploid hybrids, improving the crossability of recalcitrant cross combinations using embryo rescue techniques, the development of inbred lines using anther culture and the possible application of GM technology.
My own unpublished results suggest that the triploid hybrid variety S. x stottii ‘Ashton Stott’ could be doubled with colchicine using a root immersion technique based on that performed by Morgan (1975). This and other methods, such as that carried out by Zhang et al (1997) should be tested more fully.
Research Requirements
Now for the first time the opportunity exists for researchers to test the near- market varieties produced by the EWBP. Such collaboration will provide detailed information on elite varieties prior to their commercial release. This is far more sensible than carrying out research on established varieties or older non-registered clones. The result of the latter is that by the time the experimental data is acquired the varieties are liable to be outclassed or obsolete and therefore unlikely to be planted by biomass growers.
The superior selections created by the breeding programme should form the bedrock of future disease and pest research such as the work on mixed clonal plantations. A major flaw exists in our current breeding strategy; the breeding process is aimed at identifying superior genotypes on the basis of their performance in pure
47 stands. Once superior individuals are identified we advise that they should be planted in mixtures. A high yielding clone may be completely unsuitable for a mixture because it may be too competitive or, if it is slow to establish, not competitive enough. Research is needed to identify genotypes that are high yielding but also possess characters conducive to synergism.
Other crucial questions remain unanswered. For instance we have sparse knowledge on the physiological basis of heterosis. Fundamental research is required into the physiological composition of the highest yielding varieties, such as their water use efficiencies, leaf area indices, respiration rates and concentrations of plant growth regulators.
Novel Industrial Uses
Willow varieties that are unsuitable for use as a biomass crop may have attributes that can be exploited in other novel industrial applications such as: ♦ raw materials for the forest products industry (Deka, Wong and Roy, 1992;1994), ♦ natural dyes (Stott, Bridle and Timberlake, 1971) ♦ pharmaceuticals and other biologically active compounds (Gebhardt, 1992; Meier, 1992; Julkunen-Tiitto and Meier, 1992), ♦ livestock feeding (McCabe and Barry, 1988), and ♦ pollution control (Riddell-Black, Pulford and Stewart, 1997; Edwards, 2000).
The anticipated benefits from such novel applications are numerous and include the potential to provide the willow grower with value-added products, the provision of sustainable sources of natural products, the diversification of UK arable farming, rural employment opportunities and numerous environmental benefits.
Conclusion
The European Willow Breeding Programme has achieved a great deal in its first five years. By the end of the next five years the first products of this initiative will be in the marketplace. Nevertheless, many important strategies and research initiatives need to be implemented now to ensure the long-term success of the breeding programme and the biomass industry as a whole.
ACKNOWLEDGEMENTS
The European Willow Breeding Programme is funded by a partnership comprising IACR-Long Ashton Research Station, Svalöf Weibull AB and Murray Carter. IACR receives grant-aided support from the Biotechnology and Biological Sciences Research Council of the United Kingdom.
48 References
Ares A and Gutierres L (1996) Selection of Poplar Clones for the Lower Valley of the Colorado River, Argentina. Forestry 69, No.1: 75-82.
Baudouin L, Baril C, Clément-Demange A, Leroy T and Paulin D (1997) Recurrent Selection of Tropical Tree Crops. Euphytica 96: 101-114
Bisoffi S (1990) The Development of a Breeding Strategy for Poplars. International Poplar Commission (FAO), 35th Committee Meeting, Buenos Aires 19-23 March 1990. FO:CIP:BR/90/9. 21pp.
Bisoffi S and Gullberg U (1996) Poplar Breeding and Selection Strategies. In: Biology of Populus and it’s Implications for Management and Conservation pp. 139-158. Ed. Stettler, R F, Bradshaw, H D Jr, Heilman, P E, and Hinckley, T M. NRC Research Press, Ottawa, Ontario, Canada.
Cotterill P P and Dean C A (1990) Successful Tree Breeding with Index Selection. CSIRO Division of Forestry and Forest Products. 80 pp.
Deka G C, Wong B M and Roy D N (1992) Suitability of Hybrid Willow as a source of Pulp. Journal of Wood Chemistry and Technology 12: 2: 197-211.
Deka G C, Wong B M and Roy D N (1994) Variation of Specific Gravity, Fibre Length and Cell Wall Thickness in Young Salix Clones. Journal of Wood Chemistry and Technology 14: 1: 147-158.
Dickmann D I (1985) The Ideotype Concept Applied to Forest Trees. In: Trees as Crop Plants. Ed. Cannell M G R and Jackson J E. Institute of Terrestrial Ecology, Huntingdon, England, pp. 89-91.
Dickmann D I and Keathley D E (1996) Linking Physiology, Molecular Genetic and the Populus Ideotype. In: Biology of Populus and it’s Implications for Management and Conservation pp. 139-158. Ed. Stettler R F, Bradshaw H D Jr, Heilman P E, and Hinckley T M. NRC Research Press, Ottawa, Ontario, Canada.
Edwards R R (2000) The Potential for the Use of Willow (Salix spp.) in Buffer Zones for Reducing Nitrate and Atrazine Pollution. PhD Thesis. University of Bristol. 297 pp.
Gebhardt K (1992) Basis and Methods of Breeding Pharmaceutically Valuable Willows. (in German) Holzzucht 46: 1-4: 9-14.
Hull F H (1945) Recurrent Selection for Specific Combining Ability in Corn. Journal of the American Society of Agronomy, 37: 134-145.
Jenkins M T (1940) The Segregation of Genes Affecting Yield of Grain in Maize. Journal of the American Society of Agronomy, 32: 55-63.
49 Julkunen-Tiitto R and Meier B (1992) Variation in Growth and Secondary Phenolics Among Field Cultivated Clones of Salix Myrsinifolia. Planta Medica 58: 77-80.
Kang H, Lascoux M and Gullberg G (1996) Systematic Tree Breeding. In: Tree Breeding. Ed. Mandal A K. Indian Council of Forestry Education. Jabalpur, India.
Koski V and Dickmann D I (1992) Tree Ideotype. Biomass and Bioenergy, 2: 1-6: 71- 75.
Larsson S (1997) Commercial Breeding of Willow for Short Rotation Coppice. In: Aspects of Applied Biology 49, Biomass and Energy Crops pp. 215-218. Ed. Bullard M J, Ellis R G, Heath M C, Knight J D, Lainsbury M A and Parker S R. The Association of Applied Biologists.
Lindegaard K N and Barker J H A (1997) Breeding Willows for Biomass. In: Aspects of Applied Biology 49, Biomass and Energy Crops pp. 155-162. Ed. Bullard M J, Ellis R G, Heath M C, Knight J D, Lainsbury M A and Parker S R. The Association of Applied Biologists.
McCabe S M and Barry T N (1988) Nutritive Value of Willow (Salix sp.) for sheep, goats and deer. Journal of Agricultural Science, Cambridge, 111: 1-9.
Meier B (1992) Pharmaceutical Requirements on Willow Production. (in German) Holzzucht 46: 1-4: 5-9.
Morgan W G (1975) A Technique for the Production of Polyploids in Grasses. Euphytica 25: 443-446.
Mosseler A (1990) Hybrid Performance and Species Crossability Relationships in Willow (Salix). Canadian Journal of Botany 68: 2329-2338.
Nilsson G (1988) Willow Breeding Program at Svalöf AB. International Energy Agency Proceedings from Willow Breeding Symposium August 31-September 1, 1987, pp. 25-27. Research Notes 41, Swedish University of Agricultural Sciences, Uppsala, Sweden.
Riddell-Black D, Pulford I D and Stewart C (1997) Clonal Variation in Heavy Metal Uptake by Willow. In: Aspects of Applied Biology 49, Biomass and Energy Crops pp. 155-162. Ed. Bullard M J, Ellis R G, Heath M C, Knight J D, Lainsbury M A and Parker S R. The Association of Applied Biologists.
Stettler R F, Zsuffa L, and Wu R (1996) The Role of Hybridisation in the Genetic Manipulation of Populus. In: Biology of Populus and it’s Implications for Management and Conservation pp. 139-158. Ed. Stettler R F, Bradshaw H D Jr, Heilman P E, and Hinckley T M. NRC Research Press, Ottawa, Ontario, Canada.
Stott K G, Bridle P and Timberlake C F (1971) Anthocyanins in Willow Bark. Long Ashton Research Station Annual Report. University of Bristol, 159-160, 171-173.
50 Zhang Z, Li F, Zhu Z and Kang X (1997) Doubling Technology of Pollen Chromosome of Populus tomentosa and its Hybrid. Journal of Beijing Forestry University (English Ed.), 6: 2: 9-20.
51 Figure 1: The willow breeding scheme followed by the European Willow Breeding Programme (Elaborated from Nilsson (1988) and P1 x P2 Larsson (1997). The cycle from initial cross to registered variety may take as long as 15 years but genotypes showing superior growth can be fast-tracked through the system, reaching the marketplace in 8-10 years. Year 1 Seedling Nursery
5-10% selected
Observation Trial 1 Years 2-5 5-10% selected
Observation Observation Trial 2 Trial 2 Years 6-9 5-10% selected
1 2 3 4
Multi-site Multipn Plantations Yield Trials Year 10-15
1-2 varieties selected
Registration and Release
of New Variety52 igure 2: Progress of the European Willow Breeding Programme from 1996-2000. Already over 50 varieties are in advanced observation and yield trials. Crosses Nursery Obs I Obs II Yield
1996 107 6530 Selection
256 11602 514+60 18
1997 Selection
245 9037 775 /
1998 Selection
98 7593 50125 3
1999 Selection
2000 78 7812 504 18 2
Commercial Production and Release 53 Figure 3: Pathways used in the crossing programme 1996-2000: intraspecific crosses (a); interspecific crosses producing F1 hybrids (b); full-sib crosses producing F2 hybrids (c); first generation backcross (d); second generation backcross (e); three-way species cross (f); four-way species cross or double cross (g). (Based on Figure 2, page 100, from Stettler, Zsuffa and Wu,1996). Species Species Species Species
a a a a ABCD x x b b d f FF e g
Backcros c 3-way 4-way hybrid F (double
Backcross 54 Figure 4: Reciprocal Recurrent Selection for General Hybridising Ability (RRS-GHA). This breeding strategy involves the phenotypic selection of suitable parent plants from two populations’ (a) followed by interspecific hybridisation (b). The F1 families are tested to identify progeny with high GHA (c) and elite lines are selected and advanced through the breeding scheme possibly leading to new varieties (d). F1 hybrids with high GHA undergo full sib mating (e) and the progeny are selected on the basis of phenotype (f). The process is then repeated several times until there is no further genetic gain. (Based on Figure 1, page 99, from Stettler, Zsuffa and Wu,1996).
Species A Species B a a A/ B/ b
F1 c e d New varieties A1 B1 f A1/ B1/
F1
New varieties
A2 55 B2 56 Prof. Theo Verwijst Head of Department/Chair of Short Rotation Forestry Department of Short Rotation Forestry Swedish University of Agricultural Sciences P.O. Box 7016 SE-750 07 Uppsala, Sweden E-mail: [email protected]
BREEDING, PEST & DISEASE CONTROL: AN EUROPEAN VIEWPOINT
The relation between plant breeding on the one hand and pest and disease control on the other hand is particularly strong when dealing with genera consisting of species that hybridise easily, because such genera in general attract many different kinds of insects and pathogens. Both Salix (willow) and Populus (poplar) are genera in which hybridisation is common. Biomass yield, for which the major selection in willow breeding is performed, is determined by interactions between plants and an environment that consists of biotic and abiotic factors. Abiotic conditions vary through time, and plants may to a certain degree be well adapted to a specified range of abiotic conditions. However, pests and diseases are biotic factors that also influence on plant performance and these may not only vary in time and space but, in contrast to abiotic factors, also have the ability to co-evolve with plants. Potentially pathogenic organisms may go unnoticed for longer periods in a willow stand, until a virulent strain develops.
Susceptibility to pathogens in poplars and willows is, for certain groups of pathogens, highly clone specific. This has a number of implications for willow breeding because willow agronomy today largely is based on the use of monoclonal stands.
Willow is a boreal-arctic genus and distributed over the entire northern hemisphere, but some indigenous willow species can be found in Japan, India, Africa and Central and South America. Willow has been introduced in Australasia and in New Zealand, and for many of the species, material has been transferred far beyond its natural range. For instance, the commercial willow energy forestry systems currently in use in Sweden are based on Salix viminalis introduced during the 16th century from continental Europe for the purpose of basket making, and on hybrids between S.viminalis and some other willow species (S. burjatica, S. schwerinii), that have been introduced from Siberia more recently. Willow also is a complex taxonomical genus. Depending on the taxonomic school (lumpers or splitters) we find about 350 to 500 willow species world-wide, which is roughly 10 times as much as in Populus. It has been know for a long time that ‘Once you grow poplar, diseases will come’. The same is true for willow. Populus is a genus very closely related to Salix, and furthermore, there is relatively broad experience of breeding and pest- and disease control in poplar. Consequently, we may learn a lot from the existing poplar experience.
The world-wide transfer of willow material has consequences in the same way as we know it has for poplar. Locally, plant resistance against pathogens is much higher in hybrids because these represent a chemical composition that does not seem to be attractive or manageable for the local set of pathogens that has evolved together with the local species. However, in the long run, pathogens will co-evolve and may have devastating effects on monoclonal stands. Especially pathogens with a short generation time may develop virulent pathotypes rapidly and current trends in poplar breeding have almost given up the concept of breeding for complete disease resistance. Instead, breeding for tolerance - for being able to perform relatively well despite a certain level of pathogen infestation - is aimed at in modern breeding programmes. We have seen the result of poplar breeding strategies in Western Europe: Using a very narrow genetic base in the operational clones, many stands became wiped-out by leaf rust varieties that managed to adapt to the poplar clones. A clone that is resistant today will be susceptible tomorrow, and consequently we may better target a broad tolerance in our willow clones, using breeding programs that continuously supply us with new clones. Two other important experiences have been gained by the world-wide transfer of willow and poplar material: While moving European clones to the USA, it was discovered that some very well performing clones performed relatively poorly in the USA because the presence of pathogens (insects) that do not live in Europe. Especially in poplar cultivation we also have seen introductions of pathogens to continents where these pathogens did not occur
Page 57 previously. In poplar cultivation, special rules have been designed to avoid such spread of pathogens. In willow breeding, measures will have to be taken to diminish risk for disease spreading, and preservation of natural populations will be an issue in the future.
Willow breeding therefore is not a national or continental issue, but a global issue. Willow breeding programs are ongoing in Sweden, the U.K., Belgium, Croatia and Poland, but also outside Europe in New Zealand, Chile and in the United States. World-wide exchange of germplasm is for the good or the worse of our national programs: It may give good traits in hybrids, but is also increasing the risks of disease spreading.
If we ask willow breeders about breeding goals, we definitely get the answer that breeding is primarily towards yield increase. Quality aspects are beginning to be considered for shrub-formed willows. In most willow breeding programs an early screening for leaf rust resistance (Melampsora epitea) is implemented. One of the problems with leaf rust is the interaction with winter frost: No proper onset of dormancy is reached, leading to freezing during winter and a high mortality. During recent years resistance, against several species of leaf beetles has come into focus. Willows are known to be the host for many insect species but few studies have been performed to quantify the effect of defoliation by insects on stem wood production. The few figures from the literature indicate a production loss between 20% and 40% after severe defoliation. The main problem is though to understand the population dynamics of the actual insects to be able to judge if and how contra-measures should be taken. Slightly more than 10 years ago, problems commenced with leaf beetles in commercial plantations in Sweden. While some plantations became entirely defoliated, a student was assigned to the problem. Next spring this student had to travel thousands of miles to collect a handful of leaf-beetles and had to set up laboratory experiments to get material for his thesis. Five years later, around the time of the defence of the thesis, we again started to get extensive leaf beetle outbreaks. In Sweden we have observed completely defoliated fields adjacent to fields of the same clone where hardly any insect damages could be found. There are good indications for a control of leaf beetle populations by generalist predators, and that provision of key habitats for those predators close to willow stands is a key factor in the control of leaf beetles. In other words, it is time for integrated pest management schemes as we see for many other agricultural crops.
The progress made by current willow breeding programmes has to be integrated with growing systems design (practical planting practice) and practical management of Willow-SRC. We see in Sweden that harvest frequency recommendations have to be changed due to the fact that the newly released clones produce much faster than the ones released only 10 years ago. In other words, there is need for a continuous exchange of information between willow breeders and those actors who are running the later parts of the chain. If we consider the yield capacity of a willow clone over a period of 25 years, we find that that the top performers during good years usually have a below average performance over many years. Consequently, characteristics such as drought resistance and water use efficiency are now coming into focus as new breeding criteria. Another relevant question is if we want to target the energy market only, or if we want our willow breeding programmes to produce clones that are suitable for vegetation filter purposes or pulpwood production. If new markets will be addressed, other breeding criteria will come into focus.
The use of more or less random multi-clonal mixtures, which indeed may delay the spreading of disease, has been discussed for almost a decade and promising experimental work on mixtures has been carried out in several places. Such systems seem to work at least during shorter periods. However, it is a major challenge to compose a sustainable clone mixture in which the different clones do not out-compete each other during a time span covering several cutting cycles. The development of new molecular methods in breeding is likely to speed up the process of clone selection. Although those methods are unlikely to prevent pathogens from co-evolving with the new clonal material, they may give us - in the longer run - a much higher annual release of marketable clones. They may come to provide us with a broad genetic base in operational plantations and a variety in our willow plantations that is less prone to pests and diseases than the current willow stands.
Page 58 Yield Models for Short Rotation Coppice of Poplar and Willow: Insect and Disease Surveys 1999
Alan Armstrong, Mark Broadmeadow, Sam Evans, Paul Henshall, David Lonsdale, Robert Matthews, Nigel Straw and Ian Tubby.
Forest Research, Alice Holt Lodge, Wrecclesham, Farnham, Surrey, GU10 4LH
Introduction
Surveys of insects, disease and damage caused by other agents were repeated in all the SRC plots in 1999, using the standard protocol employed in previous years. The early season survey was carried out over the period 24 May-21 July, and the end-of-season survey during 1 September–19 October.
All plots at Phase I sites, except at Balbirnie, were harvested during winter 1998-1999 and were in their first season of the new rotation in 1999. Plots at Phase II sites were due to be harvested in winter 1999-2000, and were in their third rotation year at the time of the surveys. The differences in shoot age and structure between Phase I and Phase II sites were reflected in some of the results of the insect and disease surveys. Plots at Balbirnie had grown less well than at other Phase I sites and were being left for an extra year before being harvested. Consequently, survey results from Balbirnie have been included as part of the Phase II site data in the analyses.
The overall incidence of insect infestation and disease across all SRC plots are summarised in Tables 1-4. The Tables give the overall average of the plot mean scores for each damage category, for the extensively planted poplar and willow varieties at Phase I and Phase II sites separately. The most important categories of damage were the same as in previous years, namely:
• leaf rust caused by Melampsora spp. • defoliation by insects • diseases characterised by spots and blotches on leaves.
In addition, leaf distortion and chlorosis were recorded frequently as symptoms of severe rust infection and other leaf diseases, although a small amount of chlorosis was also noted in association with herbicide damage in a few Phase I plots. Shoot dieback was evident at 25-45% of Phase II sites, mainly following heavy rust infection in 1998, and large numbers of aphids (Tuberolachnus salignus) were found again on willow stems at some of the sites that were infested with willow aphid in 1998. Frost caused significant damage early in the season at Bonython in the extreme south-west.
Leaf rust (Melampsora spp.)
Rust infection amongst the extensive poplar varieties ‘Beaupré’, ‘Ghoy’ and ‘Trichobel’ increased at Phase II sites in 1999, compared with 1998, but decreased at Phase I sites (Figure 1). Harvesting the Phase I sites appears to have reduced subsequent infection, although not substantially. Severe rust infection still occurred at some Phase I sites, notably at Trefeinon, Bore Place and Bigbrook Farm (where mean RUST-L plot scores for ‘Beaupré’ were 4.7, 4,4 and 4.0, respectively). Across all sites, 85-92% of plots of the extensive poplar varieties recorded rust in September-October 1999 (66-84% in Sept.-Oct. 1998).
Page 59 Rust infection of the extensive willow varieties showed a similar pattern to previous years, with ‘Germany’ and ‘Q83’ infected severely and ‘Jorunn’ recording little rust (Figure 2). ‘Germany’ and ‘Q83’ recorded higher rates of rust infection in 1999 than in 1998, at Phase I and Phase II sites. There was no apparent effect of harvesting on subsequent rust incidence amongst the willow plots, and average infection rates have increased progressively over the course of the trials (Figure 2). Average rust infection of ‘Germany’ and ‘Q83’ was higher at the Phase II sites than at the Phase I sites in 1998 and 1999. This difference would appear to reflect generally higher rates of rust infection in the areas where the Phase II plots are located, rather than an influence of differences in rotation age. Across all sites, 80-88% of plots of the extensively planted willow varieties recorded rust in September-October 1999 (70 –81% in Sept.-Oct. 1998).
Geographic distribution of rust on poplar
Average mean plot scores for rust infection (RUST-L) for the extensive poplar varieties at each site are illustrated in Figure 3 (9 plots/site). The relative susceptibility of the poplar varieties to rust was similar to previous years (‘Beaupré’ > ‘Ghoy’ > ‘Trichobel’, Figure 1), and was consistent between sites, such that plot scores for the different varieties could be combined to give an overall picture of rust distribution.
The distribution of rust in 1999 was similar to that observed in 1998, but average scores were generally higher. Heavily infected sites were Dunnington (Yorks), Mawdesley (Lancs), Charity Farm and Harper Adams (E. Midlands), Ceredigion and Taironen (Wales), and Alice Holt, Loseley, Charlwood and Bigbrook Farm (S. England) (Figure 3). Loyton Bampton (near Bigbrook) would probably also have scored highly for poplar rust had not the ‘Beaupré’ plots died in the previous year, and at Bonython also, in the extreme south-west, average rust scores would have been higher had not the most susceptible stools been eliminated in the previous year.
Moderate rust infection was recorded at Sunnybrae (average RUST-L score = 2.0), which is the first time that significant poplar rust has been observed in north Scotland.
Geographic distribution of rust on willow
The distribution of rust on willows was analysed separately for ‘Germany’, ‘Jorunn’ and ‘Q83’, because of the wide differences in relative susceptibility of the varieties (Figures 4-6). ‘Jorunn’ showed very low rates of rust infection (Figures 2, 5), as observed in previous years.
Leaf rust was most widespread on ‘Germany’, following the pattern seen in 1997 and 1998, and severe infection (average score > 4.0) was recorded at sites in north Scotland, N. Ireland, Yorkshire, the east Midland, south Wales and south and south-west England (Figure 4). Many of these sites recorded equally high rates of rust infection on ‘Germany’ in 1998.
The distribution of rust on ‘Q83’ in 1999 was very similar to that recorded in 1998. High rates of infection were recorded for many sites in England and Wales, but infection rates were low in N. Ireland and Scotland (Figure 6). Severe rust infection on ‘Q83’ (scores > 4.0) was recorded at Dunnington (Yorks), Delamere (Cheshire), Soham (Cambs) and at Bigbrook Farm and Loyton Bampton in the south-west.
Page 60 Rust infection on poplars at the Intensive sites
Poplar varieties at the Intensive sites showed the same patterns of relative susceptibility to rust as were observed in 1997 and 1998. Amongst the Intensive sites in southern England and S. Wales, rust was most prevalent at Alice Holt, Loyton Bampton and Trefeinon, and at these sites the TD/DT varieties were heavily infected, the DN varieties were moderately infected, and the T varieties were least infected (Table 5). The main exception to this pattern was the continuing resistance of varieties ‘690394’ (‘Hoogvorst’) and ‘690386’ (‘Hazendans’) at Trumpington and Trefeinon. The resistance of these two varieties to rust disappeared at Loyton Bampton in 1998, and at Alice Holt in 1999. Rust infection on ‘690394’ and ‘690386’ was so severe on these and the other TD/DT varieties (except ‘Raspalje’) at Loyton Bampton in 1998 that, in combination with damage from frost, the stools died before assessments started in 1999.
Relatively high rates of rust infection occurred on ‘TT32’ at Alice Holt in 1999, as was observed in 1998.
Rust infection was less severe at the northern Intensive sites and differences in susceptibility between varieties were less apparent (Table 5). However, the new Belgian varieties ‘690394’ and ‘690386’ retained their high degree of resistance and recorded very low rates of rust infection. The DN varieties ‘Ghoy’, ‘Gaver’ and ‘Gibecq’ also recorded very little at rust at Balbirnie (E. Scotland) and Thorpe Thewles (Yorks).
Rust infection on willows at the Intensive sites
The rank ordering of willow varieties at the southern Intensive sites in terms of susceptibility to rust was very similar to that observed in 1997 and 1998. Highest rust scores were recorded for ‘Germany’, the S. burjatica x viminalis varieties ‘Stott 10’ and ‘Stott 11’, and the S. triandra x viminalis varieties ‘Q83’ and ‘ST248155’ (Table 6). Pure S. viminalis x viminalis varieties showed moderate amounts of rust infection, whereas hybrids between S. viminalis and species other than S. triandra and S. burjatica recorded little rust (Table 6). ‘Spaethii’ (S. spaethii) developed a moderate amount of rust, and ‘Bebbiana’ (S. sitchensis) was almost free of infection.
Overall rates of rust infection were lower at the northern Intensive sites and differences between the most and least affected varieties were less marked. ‘Germany’ recorded the highest rust scores at Thorpe Thewles and Loughgall. ‘Q83’, ‘ST248155’, ‘Stott 10’ and ‘Spaethii’ were also moderately infected at Loughgall.
Differences between pure and mixed plots
Rust infection in the pure and mixed plots showed a similar pattern to that reported in 1997 and 1998. The rate of infection on varieties that were less affected by rust in pure plots tended to increase when the variety was planted in mixture with varieties susceptible to rust. Consequently, because the amount of rust on susceptible varieties (eg ‘Beaupré’, ‘Germany’ and ‘Q83’) was similar in pure and mixed plots, the overall mean rust scores for the mixed plots were higher than for the average of the pure plots.
Page 61 Leaf diseases other than rust (‘spots and blotches’)
Symptoms of diseases that caused spot and blotch damage to leaves were more developed by the time of the autumn survey. On average, 6.8-12.5% of the leaf area was affected by September- October 1999 (Tables 1-4). Disease was particularly noticeable at certain sites (Table 7). Infection causing more than a 20% loss of leaf area was recorded in 18% of the poplar plots and 8% of the willow plots by September 1999. These proportions were lower than in 1998 (cf. 25% for poplars, 14% for willows).
Maximum rates of damage reached 39-41% leaf-area-lost amongst plots at Phase I sites and 64- 83% leaf-area-lost amongst plots at Phase II willow sites (Table 7). High rates of damage occurred on all of the extensive poplar varieties, but most of the high rates recorded for willows occurred on ‘Germany’. The higher maximum rates of damage recorded at Phase II sites, and greater number of Phase II plots with high scores (Table 7), indicates a greater development of leaf diseases other than rust in the taller, denser crops before cutting back. This is a reversal of the situation observed in 1998:
Percentage of plots with >20% leaf-area-lost because of disease spots and blotches
1998 1999
Phase I Phase II Phase I Phase II
Poplars 30.2 15.5 9.0 30.0 Willows 21.9 12.8 6.3 8.7
The majority of high plot scores for spot and blotch damage occurred in England and Wales south of Yorkshire. Within this area there was no geographical pattern in disease incidence.
TAPHRINA AUREA ON POPLAR
As in previous years, T aurea infection was restricted to the poplar variety ‘Ghoy’, with small amounts also occurring on ’Gaver’, both DN (P. deltoides x nigra) hybrids. Higher rates of infection were observed in the June-July survey, because infected leaves tended to fall before the autumn.
Infection rates at Phase II sites were similar to rates observed in 1998, but infection at Phase I sites was lower. All of the ‘Ghoy’ plots at Phase II sites recorded some T. aurea, and many plots recorded >10% leaf area affected. The highest mean plot scores for T. aurea occurred at Mawdesley (10-39% leaf area affected) and Delamere (8-34% leaf area affected), both Phase II sites. In contrast, only 10% of Phase I plots of ‘Ghoy’ were infected with T. aurea and in none of these was more than 10% of the leaf area damaged. In 1998, Phase I and Phase II sites showed equivalent rates of infection. The reduction in infection at Phase I sites indicates an adverse effect of harvesting on subsequent development of T. aurea or less suitable conditions for development of T. aurea in the early stages of the coppice cycle.
Page 62 There were no significant differences in the rate of infection with T. aurea between pure and mixed plots of ‘Ghoy’.
Defoliation by insects
Average rates of defoliation by leaf-chewing and skeletonising insects across all plots were typically low (8-10%, Tables 1-4), but a small number of plots were more severely affected. Between 2% and 3% of poplar and willow plots recorded defoliation rates greater than 20%, which was similar to 1998, but maximum amounts of defoliation recorded were lower than in previous years (Tables 8 and 9). The highest rates of defoliation occurred in June-July amongst poplars at Alice Holt and Llandovery 18, and amongst willows at Charity Farm and Balbirnie (Table 8).
The damage at Alice Holt and Llandovery 18 represent increases from 1998. At Alice Holt, only ‘690386’ was heavily defoliated in 1998 (by 23-48%), whereas a much wider range of varieties were affected in 1999 (Table 8). Defoliation at Alice Holt was caused primarily by brassy beetle (Phratora spp.). At Llandovery 18, the three plots of ‘Beaupré were defoliated by 33% in 1998, but ‘Ghoy’ and ‘Trichobel’ were much less affected. Alice Holt and Llandovery 18 are Phase II sites and the increase in defoliation runs parallel with the increase in growth of the crop.
The highest rates of defoliation in June-July 1999, and the greatest number of plots with >20% defoliation, occurred at Phase II sites, rather than at Phase I sites (Table 8), which suggests that populations of Phratora and other defoliating insects were adversely affected by the reduction in shoot material available after harvesting. At most sites, rates of defoliation were too low, and examples of heavy defoliation too scarce, to determine whether defoliation rates were reduced significantly after harvesting. However, defoliation of ‘Jorunn’ at Aller Court, which reached 70-80% in 1997 and 1998, during the second and third years of the crop rotation, was reduced to less than 15% in June-July 1999. Defoliation increased to 24-32% by September (Table 9). Surveys in 2000 should indicate whether defoliation at Aller Court is truly increasing again, and whether there is a correlation between defoliation rate and crop development.
Aphids infesting stems (Tuberolachnus salignus)
Infestations of aphids on stems are recorded primarily during the September-October survey, and mainly affect willows. In most cases, the infestations can be identified as colonies of the large willow aphid, Tuberolachnus salignus L.. Infestations have been restricted to a small number of sites each year, but at some of these sites the aphids have been extremely abundant. In 1999, 13 sites in England and Wales, and Castlearchdale in N. Ireland, recorded plot mean scores >1.0 for aphids on willow stems (Table 10). This compares with 4 sites in 1996, 2 sites in 1997 and 8 sites in 1998.
Aphid infestations in 1999 were most severe at Llanrwst , Aller Court and Long Ashton. Aphids were also present on willow stems at these sites in 1998, particularly at Llanrwst and Aller Court. The infestation at Llanrwst led to the death of two of the ‘Germany’ plots before the assessments in 1999. The data provide strong evidence that populations of willow aphids over- wintered at these sites. The population at Aller Court does not appear to have been affected by harvesting during the intervening winter. At this site, all of the aphids occurred on the new green stems in 1999, whereas at the Phase II sites the aphids occurred predominately on the older 2- year stem.
Page 63 No significant differences could be detected in aphid populations between pure and mixed willow plots. Apart from the death of ‘Germany’ and some dieback on ‘Q83’ at Llanrwst, heavy infestations of stem aphids in 1998 were not associated with shoot dieback in 1999.
Very few aphids (Pterocomma spp.) were recorded on stems of poplar in 1999, following the pattern of previous years. Moderate numbers occurred only at Loseley (max. plot scores 1.0- 2.5).
Frost damage and shoot dieback
Very little frost damage was observed in 1999, and only in the June-July survey. Light damage was recorded on the poplars and willows at Balbirnie, on poplars at Alice Holt and on willows at Bonython. One plot of ‘Ghoy’ was severely damaged (plot mean score = 4.1) at Bonython.
Shoot dieback was recorded in 45% of poplar plots and 41% of willow plots at Phase II sites in 1999, but in less than 5% of plots at Phase I sites which had been cut back during the winter. Serious dieback (plot mean scores >3.0) at Phase II sites occurred almost exclusively in plots of TD and DT poplar varieties that had suffered severe rust infection in 1998, especially at Alice Holt, Ceredigion, Charity Farm, Delamere, Llandovery 16, Loseley and Mawdesley. Similarly, dieback was also evident on ‘Germany’ at Ceredigion and Mawsdesley, and on this variety and ‘Q83’ at Delamere. However, dieback on willows was not as severe as on poplars.
Dieback amongst the poplars at Ceredigion might not be associated solely with previous rust infection, however, as ‘Trichobel’ at this site also recorded moderate (2.0-3.0) shoot dieback but was not being heavily infested with rust in the previous autumn.
Moderate shoot dieback noted previously on the willow varieties ‘Dasyclados’ and ‘V789’, which is not associated with rust, was observed again in 1999 at Loyton Bampton and Balbirnie.
Page 64 TABLE 1. Overall plot mean scores for each damage category for ‘Beaupré’, ‘Ghoy’ and
‘Trichobel’ at Phase I sites in 1999. (n=201 plots)
June-July 1999 September-October 1999
Overall Max plot % plots Overall Max plot % plots mean score affected mean score affected plot plot score score
(1) Incidence scores rust – leaves1 0.01 0.33 5% 1.07 4.78 85% rust – stools2 0.01 1.00 5% 1.06 3.00 82% distortion2 0.26 1.39 75% 0.21 1.56 52% chlorosis2 0.07 1.71 39% 0.16 3.00 51% stem lesions1 0.04 1.00 19% 0.18 2.35 27% leaf hoppers1 0.02 0.43 20% 0.06 1.96 21% hail damage1 0.01 0.17 16% 0.02 0.83 11% terminal bud death3 0 0.06 2% 0.01 0.35 7% aphids – leaves1 0.01 0.13 12% 0.01 0.13 11% aphids – 1 year stem1 0 - - 0.01 0.72 2% leaf galls 1 0.01 0.24 10% 0 0.11 7% shoot dieback1 0 0.11 1% 0 0.22 1% stem borers1 0 0.11 1% 0 - - frost damage1 0 - - 0 - - spittle bugs1 0 - - 0 - - aphids – 2 year stem1 0 - - 0 - - aphids – shoot tip3 0 - - 0 - - caterpillars in tips3 0 - - 0 - -
(2) % leaf area lost (0 – 100%) disease spots & blotches 3.5% 12.9% 100% 6.9% 40.6% 100% leaf chewers 4.8% 11.9% 100% 4.7% 9.9% 100% skeletonizing 3.5% 8.7% 100% 3.7% 18.1% 100% blotch leaf mines 0.1% 1.3% 15% 0.1% 2.6% 13% linear leaf mines 0.1% 2.2% 18% 0.1% 1.5% 20% Taphrina 0.1% 4.2% 4% 0 1.9% 3%
1 incidence scored 0-5 2 incidence scored 0-320% 3 incidence scored 0-1
Page 65 TABLE 2. Overall plot mean scores for each damage category for ‘Germany’, ‘Jorunn’ and ‘Q83’ at Phase I sites in 1999. (n=197 plots)
June-July September-October
Overall Max plot % plots Overall Max plot % plots mean score affected mean score affected plot plot score score
(2) Incidence scores rust – leaves1 0.09 1.98 22% 1.63 5.00 80% rust – stools2 0.16 2.00 21% 1.41 3.00 76% distortion2 0.25 1.37 60% 0.32 2.89 68% chlorosis2 0.05 1.07 28% 0.33 2.87 58% aphids – 1 year stem1 0 - - 0.20 4.44 15% stem lesions1 0.05 0.50 28% 0.10 2.00 30% terminal bud death3 0 0.11 2% 0.06 0.89 22% leaf galls1 0.02 0.31 15% 0.05 1.09 14% leaf hoppers1 0.01 0.28 10% 0.03 1.09 10% hail damage1 0.01 0.46 8% 0.01 0.44 7% shoot dieback1 0 0.50 3% 0.01 0.78 5% aphids – leaves1 0 0.20 8% 0.01 0.17 15% stem borers1 0.01 0.56 5% 0.01 0.28 4% aphids – 2 year stem1 0 - - 0 0.11 1% caterpillars in tips3 0 - - 0 0.06 1% frost damage1 0.02 1.21 2% 0 - - aphids – shoot tip3 0.01 1.00 3% 0 - - spittle bugs1 0 0.17 5% 0 - -
(2) % leaf area lost (0 – 100%) disease spots & blotches 2.6% 4.3% 100% 6.8% 39.0% 100% leaf chewers 4.6% 9.6% 100% 5.9% 22.6% 100% skeletonizing 3.3% 13.1% 100% 3.2% 10.0% 100% blotch leaf mines 0.1% 2.1% 12% 0.2% 13.9% 12% linear leaf mines 0.1% 2.4% 12% 0 1.8% 4% Taphrina 0 - - 0 - -
1 incidence scored 0-5 2 incidence scored 0-3 3 incidence scored 0-1
Page 66 TABLE 3. Overall plot mean scores for each damage category for ‘Beaupré’, ‘Ghoy’ and
‘Trichobel’ at Phase II sites in 1999. (n=227 plots)
June-July September-October
Overall Max plot % plots Overall Max plot % plots mean score affected mean score affected plot plot score score
(3) Incidence scores rust – leaves1 0.10 1.89 25% 2.29 5.00 92% rust – stools2 0.21 2.56 29% 1.77 3.00 92% distortion2 0.20 1.50 55% 0.25 2.59 60% shoot dieback1 0.57 4.94 45% 0.22 3.69 25% chlorosis 2 0.06 0.70 42% 0.12 1.31 44% leaf hoppers1 0.01 0.22 12% 0.12 2.13 16% stem lesions2 0.09 2.00 24% 0.10 1.67 22% hail damage1 0.02 0.37 15% 0.07 1.26 13% aphids – leaves1 0 0.09 10% 0.01 0.30 11% aphids – 2 year stem1 0 0.06 1% 0.01 1.00 3% terminal bud death3 0 0.06 1% 0.01 0.27 4% aphids – 1 year stem1 0 0.17 2% 0 0.50 1% stem borers1 0 0.39 6% 0 0.11 4% leaf galls1 0.04 1.15 12% 0 0.06 2% aphids – shoot tip3 0 0.11 3% 0 - - frost damage1 0.02 4.11 2% 0 - - spittle bugs1 0 0.11 1% 0 - - caterpillars in tips3 0 0.06 1% 0 - -
(2) % leaf area lost (0 – 100%) disease spots & blotches 4.0% 30.0% 100% 12.5% 82.5% 100% leaf chewers 4.7% 21.5% 100% 5.2% 15.0% 100% skeletonizing 5.7% 38.8% 100% 4.2% 22.9% 100% Taphrina 3.8% 39.0% 35% 0.5% 13.6% 18% linear leaf mines 0.1% 2.4% 15% 0.1% 1.9% 8% blotch leaf mines 0.1% 1.3% 16% 0 0.4% 2%
1 incidence scored 0-5 2 incidence scored 0-3 3 incidence scored 0-1
Page 67 TABLE 4. Overall plot mean scores for each damage category for ‘Germany’, ‘Jorunn’ and
‘Q83’ at Phase II sites in 1999. (n=237 plots)
June-July September-October
Overall Max plot % plots Overall Max plot % plots mean score affected mean score affected plot plot score score
(4) Incidence scores rust – leaves1 0.40 4.00 44% 2.12 5.00 88% rust – stools2 0.51 3.00 47% 1.59 3.00 86% aphids – 2 year stem1 0 0.28 3% 0.46 5.00 25% distortion2 0.14 1.19 60% 0.32 2.81 62% chlorosis 2 0.12 1.67 57% 0.17 2.37 37% shoot dieback1 0.33 4.39 41% 0.17 3.22 28% aphids – 1 year stem1 0 0.06 2% 0.17 5.00 8% leaf hoppers1 0.04 1.42 15% 0.11 1.57 20% stem lesions1 0.10 2.83 26% 0.11 2.17 24% leaf galls1 0.02 0.61 15% 0.04 0.83 12% terminal bud death3 0.04 0.83 14% 0.03 0.67 10% aphids – leaves1 0.01 0.39 10% 0.03 1.06 14% hail damage1 0.01 0.35 8% 0.02 0.50 8% stem borers1 0.02 1.33 8% 0.01 0.78 10% aphids – shoot tip3 0.01 0.56 5% 0.01 1.56 1% frost damage1 0.05 2.00 6% 0 - - spittle bugs1 0.01 0.44 4% 0 - - caterpillars in tips3 0 - - 0 - -
(2) % leaf area lost (0 – 100%) disease spots & blotches 3.0% 12.5% 100% 8.1% 63.8% 100% leaf chewers 5.4% 40.6% 100% 5.9% 14.7% 100% skeletonizing 3.4% 11.8% 100% 3.3% 11.9% 100% blotch leaf mines 0.1% 3.5% 11% 0 0.4% 4% linear leaf mines 0.1% 1.7% 7% 0 - - Taphrina 0 - - 0 - -
1 incidence scored 0-5 2 incidence scored 0-3 3 incidence scored 0-1
Page 68 TABLE 5. Susceptibility of poplar varieties to Melampsora rust, September 1999. (Mean RUST-L score: 0 = no rust, 5 = severe rust damage)
Variety Parent S. England S. Wales N. England Scotland N. Ireland group1
Alice Holt Loyton Trump’ton Trefeinon Thorpe Balbirnie Loughgall Thewles (3)2 (1) (1) (1) (1) (3) (1)
Beaupré TD 4.61 * 1.03 4.74 0.44 0.26 1.97 Boelare TD 3.67 * 0.90 4.78 0.59 0.21 2.44 Unal TD 3.70 * 0.86 3.46 0.22 0.13 1.86 710091 DT 4.94 * 0.78 4.37 0.88 0.09 2.56 710092 DT 4.62 * 0.77 4.49 0.19 0.48 1.70 710151 DT 4.68 * 0.86 4.20 0.72 0.13 1.78
Raspalje TD 4.43 2.72 0.80 2.92 0.16 0 1.38
6903942 TD 2.05 * 0.01 0.11 0.02 0 0 6903862 TD 3.26 * 0.01 0 0 0 0
Ghoy DN 2.99 3.55 0.67 2.87 0.10 0 1.53 Gibecq DN 2.17 1.84 0.90 3.27 0.13 0 1.52 Gaver DN 2.02 0.55 0.52 1.62 0.05 0 1.43
TT321 TB 3.47 1.98 1.93 2.71 0.65 0.66 1.42
Columbia T 2.86 1.78 0.94 1.53 0.55 0.87 1.66
Fritzi T 1.23 1.57 0.93 0.93 0.23 0.14 0.80 Trichobel T 1.46 1.30 0.67 0.83 0.29 0.59 1.13
1 Parent species: D = deltoides; N = nigra, T = trichocarpa; B = balsamifera
2 Rotation year. Phase I sites are in the first year after harvest during winter 1998-1999 (1); Phase II sites are in the third year of the rotation (3)
69 TABLE 6. Susceptibility of willow varieties to leaf rust, September 1999. (Mean RUST-L score: 0 = no rust, 5 = severe rust damage)
Variety Parent S. England S. Wales N. England Scotland N. Ireland spp.1
Alice Holt Loyton Trump’ton Trefeinon Thorpe Balbirnie Loughgall Thewles (3)2 (3) (1) (1) (1) (3) (1)
Q83 TV 3.11 4.12 3.06 1.67 1.12 0 2.26 ST248155 TV 2.88 1.37 2.69 0.82 1.49 0 2.22
Stott 10 BV 4.24 3.46 0.18 0.82 0.03 0.36 2.25 Stott 11 BV 4.23 2.55 0.74 0.99 0.41 0.08 1.16
Germany B 4.92 4.10 2.59 2.47 2.06 0.12 3.06
Spaethii Sp 1.78 2.68 1.34 1.07 0.30 0.20 2.27
Orm VV 1.21 1.79 0.73 0 0.60 0.13 0.67 Jorr VV 1.09 1.03 0.65 0.05 0.04 0.20 0.51 Jorunn VV 0.20 0.63 0.09 0 0 0.05 0.48
Ulv VV 0.22 1.33 0.21 0.03 0 0 0.42 V789 VCa 0.20 0.41 0.10 0.01 0.24 0 0.46
Delamere ACiV 0.26 0.28 0.77 0.25 1.28 0 0.23 Dasyclados CaCiV 0.03 0.33 0.32 0.09 0.85 0 0.31
Bjorn VSc 0.01 0.08 0 0 0 0 0.20 Tora VSc 0.04 0.04 0 0 0 0 0.02
Bebbiana Si 0.02 0 0 0 0 0 0
1 Parent species: A = Salix aurea; B = S. burjatica; Ca = S. caprea; Ci = S. cinerea; Sc = S. schwerinii; T = S. triandra; V = S. viminalis: Sp = S. spaethii; Si = S. sitchensis
2 Rotation year. Phase I sites are in the first year after harvest during winter 1998-1999 (1); Phase II sites are in the third year of the rotation (3).
70 TABLE 7. Sites recording >20% leaf area lost because of disease spots and blotches in September-October 1999.
Poplars Willows % leaf % leaf Site Variety1 area lost Site Variety1 area lost
Phase I sites: Phase I sites: Roves Farm Beaupré 41 Trefeinon Germany 25 - 39 Loyton Raspalje 25 Myerscough Germany 25 - 37 Gilderbeck Trichobel 21 - 25 Hayburn Germany 21 - 34 “ m - Trichobel 23 - 24 Aller Court Germany 25 - 28 Trefeinon Boelare 25 Bore Place Germany 26 Bore Place Ghoy 23 Myerscough Q83 24 Thorpe Thewles Boelare 23 “ m - Q83 21 Trefeinon 710092 21 Bore Place Jorunn 22 Gilderbeck Beaupré 20 Roves Farm Q83 21
Phase II sites: Phase II sites: Llandovery 16 Ghoy 53 - 83 Charity Farm m – Germany 70 “ m - Ghoy 35 “ Germany 64 Dunnington m - Trichobel 35 - 53 Loyton Germany 32 “ Trichobel 24 - 26 Dunstall Court Germany 32 Mawdesley m - Trichobel 27 -53 Alice Holt V789 24 – 32 “ Trichobel 34 “ Delamere 21 Carruchan Ghoy 34 - 52 “ Ulv 20 Charity Farm Ghoy 30 – 49 Bonython Germany 20 – 29 “ m - Ghoy 36 Charity Farm Q83 25 Delamere m - Trichobel 25 - 41 Mawdesley Germany 24 “ Trichobel 26 - 33 Gwent Germany 21 - 24 “ Ghoy 26 - 30 “ Q83 20 “ m - Ghoy 29 Gwent Ghoy 36 Alice Holt Columbia 38 “ 690394 33 “ Trichobel 31 “ Ghoy 22 “ Gibecq 22 “ Fritzi 22 Bonython Beaupré 31 Balbirnie 710151 30 “ 710092 27 Llangoed Ghoy 28 Long Ashton Beaupré 29 “ m - Beaupré 28 Loseley m - Trichobel 28 “ Ghoy 22 Charlwood Ghoy 26
71 Wensum m - Beaupré 21
1 m- = clone in mixture plot
72 TABLE 8. Sites recording 20% or more defoliation by leaf chewing and skeletonising insects in June-July 1999.
Poplars Willows % leaf % leaf Site Variety1 area lost Site Variety1 area lost
Phase I sites: Phase I sites: Loughgall 690394 22 Friars Court m - Jorunn 22 - 28 Columbia 21 Castlearchdale m - Jorunn 24 “ 690386 21 Friars Court Jorunn 21 “ 710151 20 Friars Court m-Ghoy 20
Phase II sites: Phase II sites: Alice Holt TT32 36 - 59 Charity Farm m - Germany 48 Llandovery 18 Beaupré 28 - 44 “ Germany 34 - 43 “ Ghoy 24 - 36 Balbrinie Bebbiana 33 Alice Holt Ghoy 23 - 35 “ Tora 28 “ Raspalje 27 - 32 “ Stott 11 26 - 28 Dunstall Court Beaupré 26 -32 “ Stott 10 23 Long Ashton Trichobel 32 “ Spaethii 22 - 23 Alice Holt Beaupré 25 - 31 “ Ulv 22 “ 690386 29 Charity Farm Jorunn 23 “ Boelare 20 - 28 Mawdesley m - Germany 23 “ 710092 22 - 27 Llanrwst Q83 21 “ 710151 26 Dunstall Court Ghoy 22 - 25 Alice Holt Gibecq 20 - 25 Llanrwst Trichobel 22 - 24 Alice Holt Columbia 22 “ Fritzi Pauley 22 “ Unal 22 “ 690394 22 “ Gaver 21 “ 710091 20 m- = clone in mixture plot
73 TABLE 9. Sites recording 20% or more defoliation by leaf chewing and skeletonising insects in September-October 1999.
Poplars Willows % leaf % leaf Site Variety1 area lost Site Variety1 area lost
Phase I sites: Phase I sites: Trefeinon Gibecq 25 – 33 Aller Court Jorunn 24 - 32 “ TT32 31 Castlearchdale m – Jorunn 22 – 27 “ 710091 26 “ m – Germany 27 “ 690386 25 Aller Court Q83 20 - 22 “ 710151 24 “ 690394 20 - 24 “ Ghoy 22 “ Raspalje 21 “ Columbia 20
Phase II sites: Phase II sites: Charity Farm Ghoy 20 – 35 Sunnybrae m – Q83 26 Llandovery 18 Ghoy 26 Balbirnie Ulv 21 – 25 Alice Holt TT32 20 - 25 Ceredigion m – Q83 22 “ Ghoy 24 Balbirnie Tora 22 “ Dasyclados 20 Mawdesley m - Germany 20
1 m- = clone in mixture plot
74 TABLE 10. Maximum plot scores for aphids (Tuberolachnus salignus) on stems of willows in September-October 1999. Sites with plot scores <1.0 not shown.
Plot scores: 0 = no aphids, 1 = a few individual aphids on stems, 2-3 = moderate infestation, 4-5 = severe infestations, ie most of stem covered with aphids. * indicates that aphids were present also in 1998
Site Willow variety
Germany Jorunn Q83 Other varieties
Phase I sites: * Aller Court 1.3 - 1.5 3.1 - 3.4 3.4 - 4.4 - * Trumpington 1.8 - 2.7 1.0 - 2.9 Friars Court 1.8 1.3 - 2.3 1.3 - 2.1 - Myerscough 2.3 - 1.0 - Castlearchdale - 1.0 - -
Phase II sites: * Llanrwst 4.0 4.0 4.0 - * Long Ashton 3.0 – 4.0 2.3 – 5.0 1.7 – 3.8 - Loseley 1.1 – 2.8 1.1 – 2.3 1.0 – 1.3 - * Harper Adams - 2.8 1.1 - Charity Farm 1.7 2.5 1.0 - Delamere 1.4 – 2.3 1.2 – 2.3 1.3 – 1.7 - Dunnington 1.7 – 2.2 1.5 – 2.3 1.2 – 1.7 - Alice Holt 1.1 – 2.9 - 1.4 - 1.6 2.2 Mawdesley 1.4 - - -
(- indicates plot scores < 1.0)
Aphids were present on willow stems at Dunstall Court and Demontfort in 1998, but not in 1999.
75 76 77 78 79 80 81 82 83 84 85 Pests and Diseases in Commercial Scale Short Rotation Coppice
Barbara Hilton Fuel Supply Manager ARBRE ENERGY LIMITED
E-MAIL: [email protected] Introduction
ARBRE Energy currently has 1,100 hectares of short rotation coppice (SRC) established as part of the long-term primary fuel source for its power plant near Selby in North Yorkshire. In order to reduce both financial and carbon costs in the transportation of the fuel, all of the coppices have been planted within a 45 mile radius of the power plant. It is intended that between 1,500 to 2,000ha will be planted in total for the ARBRE 1 plant with planting completed either in 2001 or 2002.
All of the SRC planted is willow apart from two 0.5ha areas of poplar, one planted in 1996 and the other in 1997. Up to seven varieties of willow are planted in an intimate mix at each site as recommended by Long Ashton Research Centre. The majority of these varieties have Salix viminalis, the Osier willow, as parental stock and all of them are on the Forestry Commission's approved list i.e. they have been through a plant breeding programme to assess suitability for UK conditions, have acceptable resistance to or tolerance of diseases such as rusts and can produce high yields. As it is always preferable to have a planting mix as genetically diverse as possible, a number of the newest varieties are now from the following crosses: S. viminalis x schwerinii and S. viminalis x burjatica.
The first coppices planted were two 2.5ha trial plots on Yorkshire Water land in 1993. Commercial planting for ARBRE started in 1996 although farmers' interest in the crop only began in earnest in 1998 leading to 300ha being planted in 1999 and 675ha in 2000. Therefore the majority of the 1,100ha of SRC are still within the early years of their first harvest cycle.
Various pests have occurred at a number of the coppice sites to date, the majority causing damage in the establishment year as opposed to damaging the more mature willow. Where necessary and where feasible, treatment has been carried out.
Disease Occurrence
Rust
Only one site to date has been damaged by rust, Melampsora spp. In 1996 an 11ha site was planted primarily with the new varieties of willow available at the time but also trial plots of more obscure varieties were planted plus 0.5ha of poplar, Beupre and Boelare.
In 1997 the poplars suffered a severe rust infestation which in turn affected much of the adjacent willow. All leaves were shed well before the end of the growing season. Fortunately there was no long-term damage as both the poplars and willow fully recovered the following year. Any affect on yield was unknown.
Apart from this one site, rust has not been a problem within the ARBRE coppices probably due to the willow varieties used and the planting regime of intimate mixing. Spraying against rust would not be carried out due to both financial and environmental cost.
Pest Occurrence
Leatherjackets
On ARBRE coppice sites, spraying with Dursban is carried out with the post-planting weed control wherever SRC is planted on ex-grassland and long term set-aside land in order to control leatherjackets, Tipula paludosa. Leatherjacket damage becomes evident when stems are found lying on the ground following attack just below the soil surface. The automatic and early use of Dursban prevents most damage but it has been noted that if the headlands remain un-sprayed, the leatherjackets can sometimes migrate into the edges of the crop. One application of Dursban proved ineffective at one late planted site this year due to the very dry conditions both at the time of spraying and for some time afterwards. It was felt that the leatherjackets had migrated well down into the soil as it
86 became drier and, due to lack of rain, the Dursban was not washed into the soil and therefore remained ineffective at the surface. A second application was made when leatherjacket damage was noted. Leatherjacket damage only occurs during the establishment year.
Slugs
Slugs have been a problem on a very small number of wet, heavier soil sites where they can strip the leaves on young stems within a few weeks of planting. Underground damage can also occur to the newly emerging shoots. Draza pellets have been applied to one ARBRE site where the damage was serious enough to warrant their use. Slug traps, plastic covers placed randomly across the field with pellets underneath to attract and kill the slugs, have also been used at one other site to assess the number of slugs present and whether Draza application was necessary which is was not. One of the most effective and benign forms of slug control is efficient ground preparation. If the soils are properly consolidated by rolling after planting, the slugs have difficulty populating a site as they prefer to move between and under soil 'cobbles' on unconsolidated land.
Willow beetles
One severe infestation of willow beetle has occurred to date at one of the ARBRE coppices, unfortunately the same site as suffered the rust damage. Beetle damage was first noticed late in the 1998-growing season when both the blue (Phyllodecta vulgatissima) and brassy (P. vitellinae) beetles were identified. Population counts of the beetles were carried out during the spring of 1999 as recommended by The Game Conservancy Trust and when counts of 100 beetles/m-2 were recorded the decision to spray, using Permasect 25, was made. Adult willow beetles tend to over- winter under the bark of mature trees and similar habitats within a few hundred metres of coppice plantations (Sage, et al. 1999). During late winter, early spring as temperatures increase, the adult beetles move into the edges of the coppice and begin to feed before mating and then moving further into the coppice. If the population reaches critical numbers during this edge-feeding stage, approximately 100m-2, it should prove relatively easy to spray the edges of the coppice from the headlands rather than attempt to overspray the entire crop. Overspraying an entire plantation would prove extremely costly not only financially but also ecologically as the pesticides used are not specific and would damage many non-target and beneficial insects.
Ideally an orchard sprayer would have been used to spray the pesticide from the headlands laterally into the coppice. Unfortunately however, due to the proximity of a watercourse alongside the most badly affected area of the coppice, current guidelines prohibited the use of an orchard sprayer. A hand lance was used from a tractor-towed trailer which meant the pesticide was only sprayed 6m into the crop rather than the hoped for 15m. The spraying did have some effect although the population increased again towards the end of the season but no further spraying was carried out. Harvesting of the site was due during the winter of 1999/2000 but was delayed until spring 2000 for various reasons. Willow beetle numbers were exceptionally high prior to harvest although no counts were made. As a small area of two-year-old coppice, approximately 1ha in size, was to be left unharvested it was felt that, following harvesting of the main plantation, the willow beetles would migrate to the standing willow and this area could then be sprayed. Before this could be carried out however, the farmer chose to overspray the whole harvested area with pesticide.
All of the ARBRE coppice sites have been monitored for willow beetles this year and they have been found at virtually all of the sites although not in numbers high enough to warrant any action.
Rabbits
ARBRE originally fenced each coppice site in its entirety against rabbits using British Standard fencing. In an attempt to reduce overall establishment costs and as fencing was one of the two highest costs, the other being cuttings, it was decided from the 2000 planting season onwards that fencing would only be erected if and where necessary. Advice was taken from each grower as to what degree of rabbit infestation he had in the vicinity of the proposed coppice site and the location of the main rabbit harbourage. If the grower was unsure, an ex-gamekeeper who was able to assess areas where rabbit damage might occur walked the site. Rabbits do still gain access particularly on the lighter soils where they are able to dig under the buried and out-turned fence. Rabbits within a plantation e.g where warrens are present in hedges within a plantation, are the responsibility of the grower and, generally, they use gassing or lamping to control numbers or, preferably, eradicate the problem.
Other pests
87 Occasionally other pest species can cause localised damage to a particular coppice site. Only one site has been deer- fenced, one of the original 1993 sites. Deer fencing is far too expensive to use so, at one other site where deer were potentially a severe problem, sacrificial planting i.e. higher density planting, was established adjacent to the woodland which held the deer. Nowadays, growers themselves are responsible for controlling or preventing deer from damaging their coppices. Hares are able to gain easy access over rabbit fencing but, to date, they have not caused significant damage to any of the ARBRE coppices although this may not be the case in other parts of the country.
Rooks have pulled newly emerged shoots from the ground at one site when they moved from an adjacent field where they had been feeding on other crops. Geese have grazed the tops from establishing coppice where the plantation was located next to a lake where geese were breeding. Moles are prevalent within some of the coppices but damage is rare and sporadic as they generally manage to avoid the newly planted cuttings. Once the cuttings are rooted they are unlikely to be dislodged by mole activity. Cattle have broken through fencing to graze the coppice. Badgers can break through rabbit netting if it is placed across their runs and, at one site, a badger sett had been dug within the coppice but fortunately it was noticed before a harvesting machine drove over or, more likely, into it.
Recommendations
SRC is not a high value crop. Two of its many positive aspects, particularly in promoting it to the general public, are the low pesticide usage involved in its establishment and management and its ecological benefits. All pest control methods should take these points into account.
The two main pests within the ARBRE coppices have been rabbits and willow beetles although the latter has only caused significant damage at one site. In both cases the actual economic cost of the damage they cause is unknown. Rabbit presence is relatively easily identified although estimating population numbers takes more time and effort. From a developer and grower's point of view the following questions need answers:
• What is the cost of rabbit damage compared to the cost of fencing? • Is there a threshold population under which it is not necessary to fence? • Could electric fencing be used in places for short periods after planting and cut-back? • If rabbits graze young shoots, what proportion of the cuttings will survive? • What long-term effect on yields is caused by rabbit damage? • Are there willow varities which prove unpalatable to rabbits?
Willow beetle numbers can again be monitored relatively easily by random counts within the edges of coppices. Unfortunately, to decide whether to spray and also the appropriate time to spray, monitoring needs to be carried out at frequent intervals through late winter and spring. For a developer in ARBRE's current situation, with responsibility for 1,100ha possibly rising to 2,000ha, this is not feasible. Individual farmers with a 10 or 20ha site of their own may not have the time to carry out this level of monitoring effectively. Willow beetle damage is not restricted to one particular year within a harvest cycle, it can occur at any time so all coppices need to be monitored. The Game Conservancy Trust have recommended choosing sites with care to try and avoid those areas where beetle colonies are likely to be found naturally (Tucker & Sage. 1999). Most farmers tend to choose their poorer land for SRC and this is frequently adjacent to watercourses and therefore prone to flooding which precludes many other crops. Native willows are often present probably harbouring willow beetles ready to migrate into the adjacent, establishing SRC. As willow beetles are virtually ubiquitous in the ARBRE coppices, concern for the future with regard to this pest is now apparent.
As with rabbits, there are questions that need answering:
• What is the economic cost of willow beetle damage if they are left totally uncontrolled? • Would the coppice recover from a severe, uncontrolled infestation? • What is the effect on long-term yields of beetle damage? • Although counts/m-2 are simple to carry out they are time consuming, would it be possible to develop a simple visual system for assessing numbers? • Will a benign biological control ever be developed? • Could a specific insecticide be developed that will not harm other insects? • What is the cost of planting sacrificial strips of willow around an SRC site taking into account the extra costs at planting and, potentially, not harvesting these strips for spraying post-harvest?
References
88 Sage R, Fell D, Tucker K, Sotherton N. 1999. Post hibernation dispersal of three leaf-eating beetles (Coleoptera: Chrysomelidae) colonising cultivated willows and poplars. Agricultural and Forest Entomology 1, 61-70.
Tucker K, Sage R. 1999. Integrated Pest Management in Short Rotation Coppice for Energy - A Grower's Guide. The Game Conservancy Ltd, Fordingbridge, Hampshire, SP6 1EF.
89 COMMERCIAL SRC BIOMASS PLANTATIONS – WILL THEY STILL BE GOOD FOR FARMLAND WILDLIFE?
Rufus Sage and Tracey Rich The Game Conservancy Trust, Fordingbridge, Hampshire, SP6 1EF E-mail: [email protected]
ABSTRACT
There have been a several studies concerning the ecology of SRC crops in Europe over the last 10 years or so. They tend to indicate that many pre-commercial SRC plots contain an under-storey vegetation and insect fauna that from a conservation point of view is more interesting than in most other farmland crops. The crop also attracts other animals. In particular, the studies indicate that SRC usually contains many more bird species at higher densities than any other arable crop.
It is crucial however that we understand how the findings of these early studies translate into the commercial situation. The imperative to produce as much woody biomass as cheaply as possible may still lead to considerable differences between commercial SRC plantations and the early pre-commercial plots on which much of this work was based.
The first major biomass energy project to plant commercial SRC plantations in the UK is Project ARBRE in Yorkshire. The DTI/ETSU are funding an ecological monitoring programme in a representative sample of these sites that aims to assess the ecological impact of the ARBRE SRC plantations on wildlife, in what is primarily an arable area. In conjunction with other similar studies the project should provide general guidance on the ecological impact of future commercial SRC developments.
Over four years, wildlife surveys will be undertaken in 12 large SRC plots, all on previously cultivated ground and in 12 paired arable control plots. A year-by-year comparison of the abundance and diversity of key wildlife groups in the SRC plots compared to that in the controls can thus be made. This should allow general conclusions about the ecological impact of SRC cropping in this region and the links between wildlife potential and plantation design and management to be drawn. Results from the first year of monitoring in 2000 are summarized.
Introduction
Early studies concerning the ecology of SRC crops in Europe over the last 10 years or so provide information on the potential of this new crop type to provide habitat for farmland wildlife, especially insects, plants and birds. By measuring aspects of the environment provided by this new crop-type some of the work also provides guidance on maximising this potential and on what may limit it (Coates & Say,1999; see Sage 1998 for a review).
90 For example several studies identify differences between the types of plantations in which this work was undertaken and commercial biomass plantations. Early monitoring programmes were usually based in plots of SRC planted for research and demonstration purposes or for other small scale uses such as rods for basket making. Many of these contained small patches of different age-classes and varieties and some were managed without biomass production in mind at all. It is therefore important that we understand how the findings of these early studies relating to wildlife use translate into the commercial situation where regular pesticide applications in large, mainly single age class plantations, containing a handful of varieties are likely.
Project ARBRE in Yorkshire is the first major commercial SRC development in the UK. They have now planted several hundred hectares of SRC within a 30-mile radius of their power station. This paper describes the design stage of a new four-year DTI/ETSU funded ecological monitoring programme in a representative sample of the ARBRE sites, that aims to bridge this gap in our knowledge of the ecolgical impact of this new crop (Anon 2000). First however, the paper summaries our existing knowledge of the potential for SRC crops to provide habitat for new and existing wildlife on farmland, and highlights some of the differences that commercial plots may have on particular wildlife groups
Wildlife in pre-commercial SRC plantations
Plants
Once an SRC crop is established, because weeds are relatively difficult to control and because they may be less threatening to the crops potential than in, say, a cereal, there could be scope for these crops to tolerate the presence of a ground flora. The likelihood and agronomic consequences of this are discussed elsewhere (e.g. Sage 1999, Sage & Tucker 1998). In terms of biodiversity, a ground flora would clearly be a good thing. Certainly, many of the early SRC plots in which ground flora surveys were undertaken contained an abundance of weeds or other plants, despite the shady environment (Gustaffson 1988; Sage 1995; Sage & Tucker 1998; Coates & Say 1999). Most of these studies also identified a process of change in the composition of the flora as stable perennial species replace the invasive ruderal species which characterise other crops and young SRC plantings.
There is therefore considerable evidence from the UK and Sweden that in many pre- commercial SRC plots an understorey vegetation has been tolerated and that over time, from a conservation point of view, it has become more interesting. A ground cover within SRC plantations would have biodiversity implications in terms of the plants themselves, insects and birds and possibly other wildlife groups. There is relatively little information on the interaction between weeds and SRC crops in established plantations but one study indicates a reasonably high tolerance of well-established coppice to weeds (Sage 1999). Would commercial plots display similar characteristics in the nature and development of the ground flora to these pre-commercial plots or would a desire to maintain weed free conditions and the continued invasions of typical farmland weeds prevail?
Insects
In terms of general insect biodiversity of tree species, Kennedy & Southwood (1984), who compared insect and spider species on British trees found more on willow than any other tree
91 or genus, suggesting that invertebrate diversity in SRC crops, at least in the canopy, could be high. Although numbers have been substantially lower in surveys of the canopy of pre commercial SRC plots these studies do suggest that the diversity in native willows may in part be reflected in SRC crops (Sage & Tucker 1997; Coates & Say 1999). Ground fauna surveys in SRC crops identify a community change in predatory insects with no increase in overall species (Coates & Say 1999). This may mean a mixture of SRC and other cropping in a given area would lead to increase in ground invertebrate diversity through the addition of species. There is also evidence that if SRC contains a ground flora it will further increase the invertebrate biodiversity potential of this crop. A high profile insect group, butterflies will use sheltered headlands and rides in and around SRC plantations (Sage 1996).
Like most arable crops however, SRC is also prone to damage from certain insect pest species (Sage & Tucker 1998). The use of insecticides in commercial SRC plantations would clearly diminish the insect fauna. Herbicides too have the indirect effect of reducing herbivorous and associated insects by removing the non-crop plants on which some of them live. It remains unclear the extent to which insecticides will be used in commercial SRC largely because of difficulties in accessing standing coppice. In addition access headlands and rides, common in the early survey plots, may not be a common feature of commercial plantations and hence provide poor habitat for butterflies and other wildlife that use these grassy margins.
Birds
Most studies of higher-order animals in SRC crops concentrate on birds. Studies of songbirds in pre-commerical SRC indicate that bird densities and species diversity would increase considerably compared to other arable crops and improved grasslands (Kavanagh 1990; Goransson 1994; Sage & Robertson 1996; Coates & Say 1999). SRC crops share some of the habitat features of traditional forms of coppiced woodland that are considered to be attractive to wildlife, for example in terms of structure, edge habitat, periodic harvesting (Fuller & Henderson 1992). Populations of farmland hedgerow species would probably increase and some new woodland and scrub species not dependant on old or dying trees would colonise. Migrant species, particularly uncommon in studies of arable farmland (e.g. Arnold, 1983), would increase dramatically if willow SRC were grown.
The issue of scale in particular is pertinent to studies of commercial plantations as many birds use edge habitats (e.g. boundaries between woodlands and fields) and small SRC plots are ‘all edge’. It is likely therefore that larger commercial SRC plots will contain lower densities of birds overall and this needs quantifying. However it is reasonable to conclude that if planted on arable or improved grassland, commercial SRC plantations would lead to substantial increase in the overall abundance of breeding songbirds using that land. It is also likely that some birds that are dependant on open farmland habitats with low cover, would not benefit for SRC crops. Others, such as skylark and meadow pipit have however been commonly recorded from cut SRC (Sage & Robertson 1996). It would be valuable to investigate the extent to which cut SRC can provide alternative nesting and brood rearing habitat for these declining birds. Like improved farmland, unimproved farmland habitats such as hay meadows or rough pasture also tend to contain low densities of breeding birds (Lack 1992). While SRC plantings would thus lead to increases in birds if planted on these uncommon habitats, they usually also support important botanical and invertebrate communities and less common birds and consevation. Overall, a mixture of SRC cropping and other land uses will probably produce the greatest benefits in terms of bird diversity and abundance but
92 particularly large plantations may contain relatively few birds and may displace certain important species. The new study aims to address these issues.
Wildlife monitoring at project ARBRE
Background
While most of the studies outlined above suggest that new and existing farmland wildlife species would use SRC crops, as indicated most survey work has been undertaken in a small number of relatively small experimental plantations. Table 1 provides a list of the probable differences between commercial SRC plantations and early SRC plots. Would edge effects in these small plots mean that wildlife use of larger commercial plots be less? Were the small number of plots used in some of the surveys representative? Would emerging problems with pests and disease lead to increasingly intensive management methods, for example more pesticides, and hence limit the crop’s value to wildlife?
Table 1.0 Likely differences between early and commercial SRC plantations (From Anon 2000).
Previous Plots ARBRE Commercial Plots
Size Generally small, <5ha. Increasingly large, 10ha minimum planned for Project ARBRE.
Age Structure Often mixed age classes. Typically even aged.
Planting Design Often containing clonal Typically uniform plantings trials and intimate structural of 4-6 clones. differences. Agricultural Inputs Little use of fertilizers or Fertilizers and insecticides insecticides. may be used.
Land Quality Typically ex-pastoral or low Grade 1 or 2 arable land quality land. may be typical for Project ARBRE Growth Rates Many plots consisted of Novel, fast growing clones, trial, relatively slow 2-3 year rotations may be growing clones, 3-5 year possible. rotations typical.
Experimental design
Project ARBRE in Yorkshire is the first major commercial SRC development in the UK. It is based on a 10 MW wood-fired power station requiring around 2000 ha of SRC. By 2000 the project had planted several hundred hectares of SRC within a 30-mile radius of their power station. The project provides the opportunity to make an early assessment of the likely ecological impact of a commercial SRC operation in the UK and hence enable us to refine
93 earlier work on the details of wildlife use in SRC. This work is essential as SRC has the potential to become a major form of land-use in certain areas. In addition, no previous study has properly quantified the ecological impact of SRC plantations compared to the land-use it replaces as adjacent control, or before and after plots, have not been used (e.g. Coates & Say 1999, Sage & Tucker 1998). ARBRE allows the use of control plots (otherwise similar non- SRC fields nearby) in a replicated, paired comparison.
The project therefore aims to monitor appropriate flora and fauna within and around a suitable number of SRC plantations and adjacent control sites and to use this information to assess the impact on the ARBRE SRC plantations on wildlife in the area. In conjunction with other similar studies in the future the project should provide general guidance on the ecological impact of future commercial SRC developments. There is already a new project being considered that has similar aims for commercial SRC on grassland sites (ETSU Pers. Comm.).
The basic experimental design is described in the project ARBRE ecological management project (Anon 2000). The trial consists of 12 SRC plots, all on previously cultivated ground, six planted in 1998/99 and six 1999/2000. Plot sizes range from 5 – 27 hectare. All plots have similar design, all are on previously cultivated ground (not grass or other) and all plots within 25 miles of a central point (the power station). As indicated each SRC plot has a paired control plot that is as far as possible similar to the SRC plot prior to it being planted with SRC. The maximum separation between SRC & control plot is 500 m.
ARBRE uses a standard planting strategy for its SRC sites so all study plots have a similar design. Six willow varieties are used. These are Tora, used in 30 % of each plot, Jorunn 20 %, Jorr 20%, Orm 15%, Ulv 10 %, Bowles hybrid 5 % (some slight between year variation in proportions exists). These are all varieties based on the willow Salix viminalis so that crop species diversity is low. They are planted using standard double staggered rows and harvested every three-years.
Monitoring at each of the 12 ARBRE sites will be based on repeated surveys of plants, insects and birds in both the SRC and control plots. A year-by-year comparison of the wildlife in the SRC plots compared to that in the controls can thus be made. While there is scope to measure a variety of other wildlife groups, plants and insects underpin the ecology of these habitats (Sage & Tucker 1998). We suggest that of the higher taxonomic groups birds are by far the easiest and most valuable group to monitor. Dedicated butterfly surveys would be undertaken in the plot headlands (Sage et. al., 1994). Monitoring of other groups such as small mammals would require considerable additional effort and within the broad constraints of this project could not be undertaken adequately. Comprehensive data on weed and insect pest populations will be recorded incidentally.
Preliminary results
The results from the first year of monitoring provide us with some information of the effect of very young SRC plots on wildlife (Anon 2001). This report also describes refinements to and the success of the various sampling and surveying methodologies that will be used during the four-year study. There were subtle differences in the wildlife communities found in and around the recently planted and cut SRC fields (which contained relatively little willow growth, particularly in early summer) and the adjacent controls. Overall, the abundance and diversity of plants was already greater in the SRC fields, particularly the older plots planted in 1999. The insect community recorded from the young SRC shoots in several plots already
94 included the occurrence of certain chrysomelid beetles, the main defoliating pests of this crop. Eighteen butterfly species were recorded in the first year of the study with a tendency for more individuals in headlands of the SRC plots than in the headlands of the arable controls. Butterflies require sheltered areas and the young SRC would have provided this later in the summer. More bird species were recorded from the SRC fields than the adjacent controls. This included several species typical of open farmland habitats. The overall density of songbirds at 1.5 per ha of SRC is lower than in other studies of mature SRC. Although we would expect this to increase as the project progresses it is possible that the effect of scale (i.e. large plots) is already having a reducing effect on bird densities.
REFERENCES
Anon (2000) ARBRE monitoring – Ecology of Short Rotation Coppice Plantations: Agreement. ETSU B/U1/00627/00/00.
Anon (2001) ARBRE monitoring – Ecology of Short Rotation Coppice Plantations: Interim Annual Report, 2000. ETSU B/U1/00627/00/00.
Arnold, G.W. (1983) The influence of ditch and hedgerow structure, length of hedgerow and area of woodland and garden on bird numbers on farmland. Journal of Applied Ecology 20, 731-750.
Coates, A. & Say. A. (1999) Ecological Assessment of Energy Coppice. Report B/W5/00216/00/REP. ETSU, Harwell, Oxford.
Fuller, R.J. & Henderson, A.C.B. (1992) Distribution of breeding songbirds in Bradfield Woods, Suffolk, in relation to vegetation and coppice management. Bird Study 39, 73 - 88.
Gustafsson, L. (1988) Vegetation succession during the establishment phase of an energy forest on a peat bog in East-central Sweden. Scandinavian Journal of Forest Research 3, 371-385.
Göransson, G. (1994) Bird fauna of cultivated energy shrub forests at different heights. Biomass and Bioenergy 6, 49-52.
Kavanagh, B. (1990) Bird communities of two short rotation forestry plantations on cutover peatland. Irish Birds 4, 169-180.
Kennedy, C.E.J. & Southwood, T.R.E. (1984) The number of species of insects associated with British trees: a re-analysis. Journal of Animal Ecology 53, 455-478.
Lack, P. (1992) Birds on Lowland Farms. HMSO, London.
Sage, R.B. (1995) Factors affecting wild plant communities occupying short rotation coppice crops on farmland in the UK and in Eire. In: Proceedings of the Brighton Crop Protection Conference 1995 - The role of weed control in land use change, 7D, 980-985.
95 Sage, R.B. & Robertson, P.A. (1996) Factors affecting songbird communities using new short rotation coppice habitats in spring. Bird Study, 43, 201-213.
Sage, R.B. & Tucker, K. (1997) Invertebrates in the canopy of willow and poplar short rotation coppices. Aspects of Applied Biology, 49, 105-111.
Sage, R.B. (1998) Short rotation coppice for energy - towards ecological guidelines. Biomass & Bioenergy, 15, 39-47.
Sage, R.B. & Tucker, K. (1998) The distribution of Phratora vulgatissima (Coleoptera: Chrysomelidae) on cultivated willows in Britain and Ireland. European Journal of Forest Pathology, 28, 289-296.
Sage, R.B. & Tucker, K. (1998) Integrated crop management of SRC plantations to maximise crop value, wildlife benefits and other added value opportunities. ETSU B/W2/00400/00/REP. Harwell Laboratories, Oxford. 347pp.
Sage, R.B., Fell, D.A., Tucker, K. & Sotherton, N.W. (1999) Post hibernation dispersal of three leaf-eating beetles (Coleoptera: Chrysomelidae) colonising cultivated willows and poplars. Agricultural and Forest Entomology 1: 61-70.
Sage, R.B. (1999) Weed competition in willow coppice crops: The cause and extent of yield losses. Weed Research, 39, 399-411.
96