0790-B4
David W. Langor and John R. Spence 1
The high functional and biological diversity represented by arthropods demand that these organisms be considered for utility as ecological indicators of sustainable forest management, and this requires a systematic and rigorous process including selection of potential indicators; definition of the relationship between the indicator and disturbance variables; determination of the robustness of the indicator; and application of the indicator in monitoring. In Canada, the single greatest impediment to the use of arthropods as ecological indicators is the difficulty of their identification. Consequently, most work has focused on a few relatively well-known groups (e.g. epigaeic carabid and staphylinid beetles and spiders, saproxylic beetles, Lepidoptera). Many recent studies have provided baseline data about the range of natural variation and have begun to quantify arthropod responses to natural and anthropogenic disturbances in the context of manipulative experiments or in retrospective approaches. Carabid beetles are the best studied group and sufficient sets of data now exist to permit a meta-analysis of the robustness of carabids as indicators across multiple spatial scales and in terms of their representativeness of broader ecological responses to disturbances. There is good potential to incorporate arthropod indicators into monitoring programmes in Canada, but it is still premature to do so. It is necessary to complete the ecological indicator selection process to develop the science basis for inclusion of taxa in monitoring programmes. Future research should focus on completing the ecological indicator selection process for taxa under current study as this represents the best opportunities for incorporation of arthropod indicators into biomonitoring. Research should also consider other means of monitoring arthropod biodiversity by the use surrogate ecological parameters such as ecological land classification and habitat classification systems.
The overarching issue of resource sustainability preoccupies the Canadian forest sector. While there is national consensus that sustainable forest management (SFM) is a laudable goal, major challenges remain to measure progress and describe how we can ascertain when sustainability is achieved. While some desired balance among economical, environmental and sociological goals will define sustainability, the dynamic and unpredictable nature of society's value perceptions means that sustainability cannot be a static mark, and that the challenges of achieving a shifting target will eventually emerge. Nonetheless, there is demand to develop appropriate indicators to allow assessment of progress of incorporation of present social values in SFM.
The perpetration of complex, natural forest ecosystems depends upon maintenance of a diverse array of structures and functions, some of which remain poorly understood. Presently no integrative measures of 'forest ecosystem health' are widely accepted. The biological diversity of forests, nonetheless, provides the ecological capital for preserving ecosystem integrity and for allowing adaptation to anthropogenic disturbances. Thus, biodiversity has been generally recognized as a basic component of any set of ecological indicators of SFM. Early efforts have focused disproportionately on a few well-known taxa (e.g., vertebrates). The vast majority of species and ecosystem functions reside in poorly known groups; clearly these warrant substantial attention to develop ecological indicators that can provide unbiased assessments of biodiversity responses to forest management.
Arthropods, mainly insects, a relatively poorly studied group, represent 65-70% of species in forests and play significant roles in ecosystem functioning. The uses and merits of arthropods as ecological indicators have been extensively discussed (e.g., New 1995; Niemelä 1997). Most compelling among their positive features are low costs of gathering samples reflective of populations, few species undertake large-scale migrations ensuring that population changes are reliably attributed to local changes, and their sensitivity to environmental conditions allows application of biological knowledge to identify aspects of environmental change are responsible for faunal change. Herein we review progress in selection, testing and application of arthropods as ecological indicators in Canadian forests, and highlight future opportunities and challenges.
Ecological indicators help to determine impacts of disturbance (natural and anthropogenic) on biota. Development of ecological indicators is a systematic process involving several essential steps, discussed below: selection of potential indicators; generation of data to define the relationship between the indicator and the disturbance variables of interest; determination of the applicability of the putative indicator across broader scales and, if appropriate, how well it represents the response of other taxa; and application of the indicator in monitoring to predict biotic responses to disturbances.
The first step in development of suitable ecological indicators is choice of potential taxa or assemblages for study. There is no clear answer about which groups are "best"; however, effective indicators must meet two general kinds of criteria (reviewed by McGeoch 1998): economic and logistical viability (e.g., cost of sampling and processing) and biological efficacy (e.g., considerations of taxonomy, distribution, representation, sensitivity). Although arthropods have many features that make them superior as potential indicators, they have been less commonly used than vertebrates.
The single greatest limitation to the use of arthropods as ecological indicators is the difficulty of their identification. Of the approximately 60% of Canadian terrestrial arthropods known and described, adequate keys exist for only a small fraction. Thus, the first considerations in selection of arthropod taxa for study as potential indicators are the taxonomic soundness and stability and the availability of resources (keys, collections, expertise) to facilitate reliable identification. The use of higher level taxa (e.g., genus) or morphospecies (apparent species) have been advocated as a solution where taxonomic expertise is too expensive or unavailable (Oliver and Beattie 1996); however, use of such surrogates is risky. Although higher level taxa may indicate biodiversity hot spots or areas of high conservation value, they will not serve generally as useful ecological indicators for arthropods as different species in the same genus usually have different habitat requirements and, therefore, often exhibit different responses to disturbances. The usefulness of morphospecies designation in taxonomically poorly known groups depends greatly on the skill and experience of the designator. Ideally, the designator should be a taxonomist expert in related groups to minimize the risk of poor morphospecies definition and resultant misidentifications. Furthermore, to satisfy the essential scientific criterion of repeatability, extensive voucher collections should be retained.
Alternatives to authoritative species-levels identifications are rarely acceptable in the development of quality indicators. Research teams developing arthropod ecological indicators simply require considerable taxonomic expertise. This may be obtained through partnership with expert taxonomists elsewhere, but often the prospect of being inundated with thousands of specimens for identification generates reluctance for taxonomists to become involved. Furthermore, the large recent declines in biosystematics expertise and training in Canada has left a shortage of qualified taxonomists. Our own solution has been to provide taxonomic training for members of the team or to hire taxonomic expertise dedicated to the project. Training provides a good solution to the taxonomic impediment in both the short and long term, but often it requires a considerable investment of time before the trainee is sufficiently skilled. Those interested in developing ecological indicators have not yet accepted this as a cost of doing business. The usual end result of the taxonomic challenges presented by arthropods is that the selection of study taxa is strongly biased towards taxa that are taxonomically well-known and for which identification skills and tools are readily available or fairly easily learned. If our goal is respond to unwanted change by altering management, it is relatively useless and certainly economically dubious to monitor taxa that are not easily dealt with taxonomically.
In light of taxonomic challenges, work in Canada has focused on a few groups of arthropods, and these are typically among the most common groups studied world-wide in northern forests. Ground beetles (Carabidae) have received most focus as they are ecologically and taxonomically well-known, diverse, ubiquitous in most habitats, sensitive to environmental stresses in various ecosystems and are relatively easily sampled with pitfall traps (Langor et al. 1994; Niemelä et al. 1993a, b; Duchesne and McAlpine 1993; Spence et al. 1996, 1997). Two other ground-dwelling groups, rove beetles (Staphylinidae) and spiders (Araneae), are caught contemporaneously with carabids in pitfall traps. Taxonomic expertise on these groups is now more common in Canada, and recently they have also been studied as potential ecological indicators (Brumwell et al. 1998; Buddle et al. 2000; Gandhi et al. 2001). With increasing focus on the values of dead wood and recognition that insects associated with dead wood are among the most sensitive groups to forest management in northern Europe, saproxylic beetles (Coleoptera) are receiving attention as potential ecological indicators in Canada (Hammond 1997; Spence et al. 1997; Hammond et al. 2001). Among foliage-feeders, only Lepidoptera has received notable study, largely because this is one of the best known and easily sampled groups (Martel and Mauffette 1997; Morneau 2002). Finally, some groups of soil arthropods (mites, Collembola) have been examined as ecological indicators (Paquin and Coderre 1997), although these groups have received more attention as environmental indicators. Undoubtedly other taxa are potential ecological indicators, but appropriate research is currently lacking.
Once a potential indicator taxon or group has been chosen, sampling is required to understand the natural range of variation and to elucidate the relationship between the indicator(s) and selected abiotic or biotic variables. There are several important considerations relevant to sampling and data collection. First, while it is easy to collect arthropods, each trapping method and sampling protocol usually has inherent biases which most be at least recognized for correct interpretation of data (Spence and Niemelä 1993b; Hammond 1997). In theory, the problems of trapping biases are minimized as long as the same approach is used across treatments and replicates, but this assumption merits debate and examination. Ideally, it would be better to sample actual numbers of insects per unit of habitat, but the enormous investment of resources required would exclude all but work at the smallest scales. Secondly, arthropod populations vary seasonally and this variation must be either controlled for (a challenging task) or sampled continuously (or with short periodicity) throughout the entire activity period, and considered in interpretations. Thus, sampling of arthropods over a short portion of the activity season, as sometimes advocated for monitoring purposes (http://www.fmf.ab.ca/pro.html), will not provide accurate assessment of either the presence/absence or relative abundance of species within stands nor allow for meaningful comparisons over space and time, as required by monitoring goals. Thirdly, although trapping is relatively cost-effective, the processing of specimens (extraction of targeted taxa, preservation, labeling, identification) requires 10-20 times more effort than sampling. Thus, it is largely the processing costs that limit the scope of projects and determines the rate at which data are accumulated.
The first requirement for development of an indicator is baseline information about distribution and abundance or species and assemblage composition under conditions deemed to be `normal'. The goal is to understand the range of natural variation (RNV) as a model against which to compare responses to anthropogenic disturbances. We remain at the early stages of such work in Canada because our fauna is relatively poorly known. Although the importance of such faunistic work is frequently underestimated, it results in a treasure of information about composition of assemblages and species distribution (e.g., Hammond 1997). Because arthropods are highly sensitive to specific habitat characteristics and exhibit seasonality, numbers of a species in an area can vary by several orders of magnitude over a short time. Thus the RNV can be wide, and this cautions that reliance on the RNV to establish response thresholds for arthropod ecological indicators is not promising in the absence of understanding how it is affected by conditions. The indicator value of arthropods can, therefore, only be realized by examining numerical trends over longer time periods and larger areas in the context of adequate meteorological records and biological understanding. Although not widely discussed, these basic limitations doubtlessly apply to most indicators.
There has been good progress in Canada in understanding the relationship between arthropod indicators (in particular carabids) and disturbances. Some species in all groups of arthropods studied exhibit responses to disturbances; some (forest specialists) are adversely affected, others (open habitat specialists) are favored and both groups have potential as ecological indicators (Niemelä et al. 1993a, Buddle et al. 2000). Species adapted to fire (a common natural disturbance in most Canadian forests) or to fire residuals (skips) are also potentially good indicators for natural disturbance regimes (Gandhi et al. 2001). Development of arthropod indicators will depend on quantifying their response to varying degrees of disturbance, either in the context of manipulative experiments (e.g., Spence et al. 1999) or in retrospective approaches (e.g., Buddle et al. 2000). Such efforts require due consideration of scale.
Because of the limitations discussed above, individual arthropod studies are conducted at relatively small spatial scales. Clearly, it is important to ascertain whether relationships may be scaled up additively to represent the situations at larger scales. Meta-analyses of multiple datasets over increasing scales might thus be used to test the spatial robustness of indicators. For potential indicator taxa commonly studied in Canada, only work with carabids has yet yielded sufficient data to support such analyses, which are imminent. As data for other groups accumulate rapidly, similar meta-analyses will be possible in a few years.
The usefulness of an indicator also depends on how well it represents responses of other groups. Most studies on arthropod ecological indicators world-wide have progressed only so far as to develop correlative data to define the relationship between potential indicators and variables of local interest. This of course provides valuable information for the conservation of the so-called "indicator taxa" but does nothing to indicate how well responses serve as a surrogate for other taxa. There are now several studies in Canada (e.g., Spence et al. 1997) that have examined the responses of multiple taxa to disturbances. As these datasets are analyzed together, individual groups can be assessed for their capacity as ecological indicators of broader ecosystem responses.
There is much current interest in monitoring forest biodiversity. The stated purpose of monitoring is to systematically assess a suite of proven and effective ecological indicators over various spatial and temporal scales to detect incipient change in ecosystem structure, function and composition in response to natural and anthropogenic influences. To date few terrestrial arthropod ecological indicators have been used in monitoring, and the potential for their future application rests on the outcome of the ecological indicator selection process (McGeoch 1998). There is good potential to incorporate arthropod indicators into monitoring programs in Canada, but it is still premature to do so. It is necessary to complete the ecological indicator selection process to develop the science basis for inclusion of taxa in monitoring programs. Unfortunately, not all monitoring programs have embraced such rigor in choice of ecological indicators for monitoring, e.g., the proposed Alberta Forest Biodiversity Monitoring Program (http://www.fmf.ab.ca/pro.html). Clearly much effort is yet required to identify robust arthropod ecological indicators for biomonitoring, and similar challenges exist for other groups of organisms (e.g., vertebrates, plants) typically touted as ecological indicators.
Development of ecological indicators requires significant research investment to provide appropriate scientific underpinnings. In northern forests, work on development of arthropod indicators of SFM has focused on a few groups. Although other arthropod groups have potential as indicators, it is questionable whether huge investments should be made in work on new groups or in monitoring programs until the indicator value of groups under current study has been demonstrated. Research should be focused on completing the ecological indicator selection process for taxa under current study as this represents the best opportunities for incorporation of arthropod indicators into biomonitoring.
Even if good ecological indicators are identified among arthropods, there is still the question of whether it is economical to include them in extensive biomonitoring programs. It is likely that the cost of incorporating arthropod indicators will be substantial. It is, therefore, wise to consider other means of monitoring arthropod biodiversity, in particular the utility of using surrogate ecological parameters. For example, ecological land classifications (ELCs) have been developed for most of Canada and these habitat `taxonomies' are largely based on soil moisture and nutrient gradients as indicated by soil profiles and selected plants. There appears to be merit in assessing the utility of ELCs as surrogates for a large range of biodiversity (including arthropods), especially those groups with strong ecological links to soils (e.g., carabids, some staphylinids, soil meso-fauna). We also advocate research to develop habitat classification systems as surrogates for biodiversity. As one example, saproxylic insects are known to be very sensitive to forest management and have high values as ecological indicators in European forests. However, this ecological assemblage is very poorly studied in Canada, and there are major taxonomic impediments to incorporation of many saproxylic taxa into monitoring programs. A functional habitat classification system might help better ascertain which habitats are most endangered due to harvesting practices and provide forest managers with a tool to help them plan dead wood retention within harvested landscapes.
Brumwell, L.J., K.G. Craig and G.G.E. Scudder. 1998. Litter spiders and carabids beetles in successional Douglas-fir forest in British Columbia. Northw. Science 72:94-95.
Buddle, C.M., J.R. Spence and D.W. Langor. 2000. Succession of boreal forest spider assemblages following wildfire and harvesting. Ecography 23:424-36.
Duchesne, L.C. and R.S. McAlpine. 1993. Using carabids beetles as a means to investigate the effects of forestry practices on soil diversity. Forestry Canada, PNFI Tech. Rep. 16.
Gandhi, K.J.K., J.R. Spence, D.W. Langor and L.E. Morgantini. 2001. Fire residuals as habitat reserves for epigaeic beetles (Coleoptera: Carabidae, Staphylinidae). Biol. Conserv. 102:131-141.
Hammond, H.E.J. 1997. Arthropod biodiversity from Populus coarse woody material in north-central Alberta: a review of taxa and collection methods. Can. Entomol. 129:1009-1033.
Hammond, H.E.J., D.W. Langor and J.R. Spence. 2001. Early colonization of Populus wood by saproxylic beetles (Coleoptera). Can. J. For. Res. 31:1175-1183.
Langor, D.W., J.R. Spence, J. Niemelä, and H.A. Carcamo. 1994. Insect biodiversity studies in the boreal forests of Alberta, Canada. Metsäntutk. tiedonantoja 482:25-31.
Martel, J. and Y. Mauffette. 1997. Lepidopteran communities in temperate deciduous forests affected by forest decline. Oikos 78:48-56.
McGeoch, M.A. 1998. The selection, testing and application of terrestrial insects as bioindicators. Biol. Rev. 73:181-201.
Morneau, L. 2002. Partial cutting impacts on moths and lepidopteran defoliators in a boreal mixedwood forest of Alberta. M.Sc thesis, University of Alberta, 138 pp.
New, T.R. 1995. An introduction to invertebrate conservation biology. Oxford University Press, U.K., 194 pp.
Niemelä, J. 1997. Invertebrates and boreal forest management. Conserv. Biol. 11:601-610.
Niemelä, J., D.W. Langor and J.R. Spence. 1993a. Effects of clear-cut harvesting on boreal ground beetle assemblages (Coleoptera: Carabidae) in western Canada. Conserv. Biol. 7:551-561.
Niemelä, J., J.R. Spence, D.W. Langor, Y. Haila, and H. Tukia. 1993b. Logging and boreal ground-beetle assemblages on two continents: implications for conservation. pp. 29-50, In Gaston, K.J., T.R. New and M.J. Samways (eds.), Perspectives in Insect Conservation, Intercept Publishers Ltd., U.K.
Oliver, I. And A.J. Beattie. 1996. Designing a cost-effective invertebrate survey. Ecol. Appl. 6:594-607.
Paquin, P. and D. Coderre. 1997. Changes in soil macroarthropod communities in relation to forest maturation through three successional stages in the Canadian boreal forest. Oecologia 112:104-111.
Spence, J.R. and J.K. Niemelä. 1994. Sampling carabid assemblages with pitfall traps: the madness and the method. Can. Entomol. 126:881-894.
Spence, J.R., D.W. Langor, J. Niemelä, H.A. Carcamo and C.R. Currie. 1996. Northern forestry and carabids: the case for concern about old-growth species. Ann. Zool. Fenn. 33:173-184.
Spence, J.R., D.W. Langor, H.E.J. Hammond and G. Pohl. 1997. Beetle abundance and diversity in a boreal mixedwood forest. pp. 285-299 In Watt, A.D. and N.E. Stork (eds.), Forests and Insects. Chapman and Hall, London.
Spence, J.R., C.M. Buddle, K. Gandhi, D.W. Langor, W.J.A. Volney, H.E.J. Hammond and G.R. Pohl. 1999. Invertebrate biodiversity, forestry and emulation of natural disturbance: a down-to-earth perspective. pp. 80-90 In Pacific Northwest Forest and Rangeland Soil Organism Symposium. USDA Forest Service, Gen. Tech. Rep. PNW-GTR-461, Portland, OR.
1 Natural Resources Canada, Canadian Forest Service, 5320-122 Street, Edmonton, Alberta, Canada, T6H 3S5. [email protected]