John A. Muir, Donald C.E. Robinson and Brian W. Geils 1
Many insects, fungi and plants in forest ecosystems can damage trees and forests, depending on stand and environmental conditions. Natural disturbances, harvesting and other forestry practices can retard or increase the spread and effects of dwarf mistletoe (Arceuthobium tsugense) and other diseases on tree growth. Certification and monitoring programmes typically use incidence and severity of infestations as criteria and indicators, but these are often insufficient to characterize some effects and measure the sustainability of new management practices, such as variable retention silviculture. Long-term observations and models of stand development are advocated as better methods for characterizing disease effects on forest practices. For dwarf mistletoe we are designing and monitoring installations in infested stands of western hemlock (Tsuga heterophylla) and constructing a spatial and life history model of stand and disease development. Spread and impacts are affected by several factors, including site quality, stand density and spatial arrangement of infected trees that are sources of mistletoe spread into new stands. Potentially, these factors could be manipulated either to reduce or encourage the spread and effects of dwarf mistletoe.
In forest ecosystems a diverse group of fungi and plants, including dwarf mistletoes (Arceuthobium species) cause tree diseases, major damage to forest ecosystems and affect sustainability of forestry practices. These are prevalent and often damaging in old growth forests, including those in Canada and British Columbia that are reserved or managed for a variety of objectives such as timber, wildlife, recreation, and water. However, infestations can also create desired features such as dead and dying trees that provide habitat for wildlife, and canopy gaps that encourage establishment of a variety of tree species and vegetation (Lundquist and Beatty 2002). In young managed forests, it could be appropriate to encourage development of some attributes associated with disease infestations, snags and canopy gaps. The challenge is to determine what management practices could encourage or suppress infestations or their effects without unduly affecting forest productivity, and where and when such practices could be applied.
Maintaining healthy forests is a key objective of recent initiatives to develop regulations and certification programs that will ensure sustainability of forestry practices and good performance by industry and government. National forestry monitoring and reporting programs, and most programs that have been developed to certify sustainable forest practices, include indicators of forest health and ecosystem condition usually measured as occurrence (incidence) and severity of damage caused by insects and diseases (Canadian Council of Forest Ministers 2000). Although these measures are used traditionally for insect infestations, they are insufficient to characterise many effects, including stress caused by abiotic agents (Innes and Karnosky 2001). Many recent or proposed changes in forest practices could exacerbate damage from diseases and create substantial uncertainties about certification programs being developed and implemented to track and ensure sustainability of management or preservation of forest ecosystems. Unfortunately, very few data or tools are available to analyse and project effects of current or future disease infestations. To describe these issues and suggest possible solutions, we briefly review several important factors for monitoring effects of diseases, and outline an approach for modelling effects of hemlock dwarf mistletoe (Arceuthobium tsugense), a widespread parasite of western hemlock (Tsuga heterophylla) in western coastal North America.
Incidence and severity attributes are very appropriate for monitoring and reporting occurrences of defoliating insects and bark beetles. Generally, effects of these insects are relatively well known and current measures of occurrence and damage are key components for decision-making in management programs. Incidence and severity of damage by insects are readily determined by annual aerial and ground surveys, usually over extensive areas. In many instances, long-term data or observations are available to determine historical trends, recurrent cycles, and recent changes in insect disturbances.
However, for common tree diseases and/or pathogens including stem rusts and cankers, foliar diseases, root diseases, wood decay fungi, and dwarf mistletoes, measures of incidence and severity are not always adequate. Signs and symptoms of disease can be cryptic or unrelated to the potential for pathogen increase and tree damage. Many diseases develop slowly but persist, so that duration of an infection and cumulative effects are often important. There are also significant differences in impacts due to pathogen life history, mode of disease spread and intensification, and type of host tissue infected.
Foliar diseases are similar to damage by defoliating insects and directly affect current or successive annual growth. Incidence and severity of infestation are reasonably well correlated with effects on tree mortality and stand growth. Foliar disease surveys can be conducted annually in early summer to determine the extent and severity of infestation. Unfortunately, there are few or no available treatments for foliar diseases in forests except for planting trees with higher levels of genetic resistance. Although foliar diseases generally cause only limited or sporadic damage, in some instances they are extremely damaging or even lethal for some species.
Stem cankers and rusts can be well characterised by measuring incidence and severity by ground-level surveys. On small trees, infections can cause mortality within one or two years, but in older trees, infections can take several to many years to girdle and kill a tree, so that effects can include a decline in tree growth over many years. Stem disease incidence and severity are often highly variable within stands and between regions, depending on genetic resistance, weather and local factors. Stem infection persists for several years but in some instances increases or accumulates for periods of one to two decades or longer. Trees planted in ecosystem zones where a tree species does not commonly grow can be severely and repeatedly infected for many years.
Root diseases have widespread, long-term and pronounced effects on forest ecosystems, but effects are not well characterised by measures of incidence and severity. Individual infected trees and/or centres of dead and dying trees can be measured using aerial and ground surveys. However, critical aspects such as belowground infection of live roots and persistent infestation of dead tree roots are not detectable by usual operational surveys (Morrison et al. 2000). Detailed excavations and sectioning of roots and stems are necessary to determine the true extent of infection occurrence and severity of root disease, but these are not practical for forestry surveys. Root diseases can also cause substantial long-term growth reductions, but these can be confounded or obscured by other factors that affect growth such as tree age, site quality, species composition and stand density. Selective cutting can also exacerbate armillaria root disease mortality and growth losses (Morrison et al. 2001).
Measures of wood decay in live trees are not feasible for forestry surveys unless trees are felled and sectioned. Several fungi including root disease fungi can infect live trees and cause wood decay. This can lead to rejection or degrade of logs or lumber and structural weaknesses in live trees that can result in tree collapse or increased vulnerability to windthrow. Wood decay is common in old-growth forests but is generally not believed to be a problem in young stands. However, thinning or partial-cut harvesting that results in more tree wounding can exacerbate wood decay. Consequently, wood decay will likely become an increasingly important issue in both immature and mature trees, particularly in variable retention silviculture systems.
Dwarf mistletoes can be generally characterised by incidence and severity ratings (Geils et al. 2002). It is possible to detect severely infected stands by aerial surveys, but incidence and severity usually have to be determined by ground surveys or inventory sampling. In young trees, dwarf mistletoe incidence and severity can be low but effects on tree growth are slow to develop and can be highly variable, depending on stand and environmental conditions. Dwarf mistletoes can be readily controlled by clear-cut harvesting and sanitation cutting to remove all residual infected trees, but with recent initiatives to retain live trees in harvested areas and limit clear felling, dwarf mistletoe impacts can increase substantially.
Given the prevalence and wide influence of diseases and pathogens in forest ecosystems, and limitations of measures of incidence and severity to show their effects, other approaches are needed to characterise effects on sustainability of forest practices. We suggest two methods: establish experimentally designed installations in conjunction with monitoring programs to determine disease effects, and develop disease forecast models to predict impacts.
Forest tree diseases are extremely variable in their nature, occurrence and impacts. We need to determine effects of major tree diseases under selected management regimes for major tree species in each important ecosystem. An experimental design is needed to ensure representative, unbiased sampling of levels of disease occurrences and sufficient replication over a range of ecosystems. Wherever possible, installations should incorporate existing trials, experiments and long-term inventory growth plots where these have been properly selected and include sufficient measurements to characterise diseases. Designs must also include both conventional forestry practices and new regimes such as variable retention silviculture. It is evident that data will always be limited in applicability due to continuing changes in forest practices.
A second approach that we recommend for determining impacts and extending current data is to develop disease-forecast models. Models for each disease or pathogen should be linked to inventory data and tree growth models. These can be used to predict stand yields and timber supplies (e.g., Woods et al. 2000). Potentially, models could help to identify gaps in knowledge and suggest new experiments. They could also enable a determination of effects and impacts under a wide range of stand and environmental conditions at various tree ages and under different management regimes. They could also assist in visualising effects and resolving debates and differences in opinion about effects and impacts in new management regimes. However, there are relatively few available disease models, and several difficulties that hinder their development or application.
As an example, we are working on a detailed model for hemlock mistletoe in coastal western hemlock forests in British Columbia and potentially for other dwarf mistletoes in other regions (Robinson et al. 2002). The model incorporates research results from many areas in the Pacific Northwest and projects stand-level effects of dwarf mistletoe for average stand conditions. We are analysing factors that could affect spread or severity of infection because hemlock dwarf mistletoe is extremely variable in incidence and severity, presumably in response to a wide range of climatic and stand conditions. Dwarf mistletoe survival, spread and effects are substantially affected by disturbance events such as windstorms (Trummer et al. 1998) or logging, and are correlated with the numbers and spatial distribution of surviving mature infected trees.
In young regenerating stands, mistletoe seed production can be vigorous and result in extensive infection and spread of the parasite, but subsequent impacts depend on density and growth of the young stands. Where stands remain relatively open, mistletoe infection is able to keep up with tree height growth and trees become severely infected. When stands are dense and height growth is rapid -approximately 50 cm or more per year typically on medium to good sites - trees can grow faster in height than the mistletoe can spread upwards. In dense stands, mistletoe shoots and seed production are suppressed by shading, dwarf mistletoe is excluded from upper tree crowns, and impacts remain low. However, even a low incidence of infection in young stands can pose a threat of future damage after thinning or partial cutting. Individual mistletoe infections on stems or lower branches on living trees can survive in a quiescent state for several to possibly 10 or more decades, and then resume vigorous seed production after a new disturbance.
We have been incorporating these phenomena of spread in the detailed model. Several outputs support hypotheses that impacts can be strongly influenced by factors such as numbers and spatial distributions of infected trees and proportions of less- or non- susceptible tree species. We are acquiring more field measurements and long-term tree growth data to test these possibilities. We hope that a model will assist foresters and others to determine the tolerable or acceptable levels of impact of dwarf mistletoe under various conditions or practices. For example, what levels of dwarf mistletoe severity would indicate significant growth impacts? On an individual tree basis, growth reductions of 15 to 30 per cent of annual growth rates are statistically significant at a mistletoe severity rating of 3 to 4. However, an average rating for a particular stand for tolerable levels of infestation could be much higher or lower depending on several factors and management objectives.
Forest tree diseases are integral components of forest ecosystems and often have major effects both detrimental and desirable on resources and sustainability of forestry practices. Recent forest legislation for harvesting, regenerating and managing immature forests includes regulations and forestry practices to prevent or reduce current or future damage from diseases. However, current measures of incidence and severity are insufficient to characterise many disease effects and impacts on forest ecosystems. Even where measures are feasible and appropriate, data on incidence and severity must be qualified because most diseases have decreasing effects as trees grow in size, with increasing tree age, and/or at higher rates of tree growth. Generally, trees on good sites are able to tolerate more disease or damage with less effect on growth or survival. Another reason for these deficiencies is that many disease effects are inconspicuous and/or difficult to measure. Unlike insect outbreaks that occur for only one or a few years, many disease effects are manifested slowly but accumulate over several decades so that damage becomes most severe and apparent in older immature or mature stands. Forest ecosystem characteristics such as species composition and stand density affect the impacts of many diseases. Further, many diseases are substantially affected by or interact with forestry practices.
Apparently there has been little critical examination of the adequacy of proposed criteria and indicators for diseases, and we are not aware of any current research to develop better measures of their effects. Criteria should incorporate more specific measures of disease effects on long-term tree growth and ecosystem attributes. Current data on disease effects are limited. We suggest that disease effects should be determined by experimentally designed long-term measurement plots and development of growth impact models for each important disease. A model under development for hemlock dwarf mistletoe shows promise in extending the usefulness of current data to new ecosystems and predicting impacts and effects of new management regimes under a range of environmental and ecosystem features.
Canadian Council of Forest Ministers. 2000. Criteria and indicators of sustainable forest management in Canada. National status 2000. Natural Resources Canada - Canadian Forest Service Ottawa, Ontario. 122 p.
Geils, B.W., J. Cibrián Tovar and B. Moody, tech. coords. 2002. Mistletoes of North American Conifers. Gen. Tech. Rep. RMRS-GTR-98. Ogden, UT: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station. 123 p.
Innes, J.L. and D.F. Karnosky. 2001. Impacts of environmental stress on forest health: the need for more accurate indicators. Pp. 215-230. In: R.J. Raison, A.G. Brown and D.W. Flynn, eds. Criteria and Indicators for Sustainable Forest Management. CABI Publishing, Oxon, UK and Int. Union For. Res. Org., Vienna, Austria. 476 pp.
Lundquist, J.E. and J.S. Beatty. 2002. A method for characterizing and mimicking forest canopy gaps caused by different disturbances. Forest Science 43: 582-594.
Morrison, D.J., K.W. Pellow, D.J. Norris and A.F.L. Nemec. 2000. Visible versus actual incidence of armillaria root disease in juvenile coniferous stands in the southern interior of British Columbia. Can. J. For. Res. 30: 405-414.
Morrison, D.J., K.W. Pellow, A.F.L. Nemec, D.J. Norris and P. Semenoff. 2001. Effects of selective cutting on the epidemiology of armillaria root disease in the southern interior of British Columbia. Can. J. For. Res. 31: 59-70.
Robinson, D.C.E., B.W. Geils and J.A. Muir. 2002. A spatial statistical model for the spread of dwarf mistletoe within and between stands. In: N.L. Crookston and R.N. Havis, comps. 2002. Second Forest Vegetation Simulator (FVS) Conference; February 12-14, 2002, Fort Collins, CO. Proceedings RMRS-P-25. Ogden, UT: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station. p 178-185.
Trummer, L.M., P.E. Hennon, E.M. Hansen and P.S. Muir. 1998. Modeling the incidence and severity of hemlock dwarf mistletoe in 110-year-old wind-disturbed forests in southeast Alaska. Can. J. For. Res. 28: 1501-1508.
Woods, A.J., A. Nussbaum and W. Golding. 2000. Predicted impacts of hard pine stem rusts on lodgepole pine dominated stands in central British Columbia. Can. J. For. Res. 30: 1-6.
1 Forest Pathologist, BC Ministry of Forests, Forest Practices Branch, PO Box 9513 Stn Prov Govt, Victoria, British Columbia V8W 9C2, Canada. firstname.lastname@example.org, Website: http://www.for.gov.bc.ca/hfp/forsite/forsite.htm