Consortium for Risk Evaluation with Stakeholder Participation

A New Approach to Assessing Ecological Health: Developing an Index of Biological Integrity with Insects at Hanford

Diana N. Kimberling and James R. Karr, Ecological Health Task Group,
CRESP UW, Seattle, WA


Until we measure the biological responses that result from human actions, we cannot accurately predict their consequences, or their associated risks, for human society. The Hanford Nuclear Reservation in eastern Washington, an ideal site for documentation of those responses, has been subjected to diverse human activities. A substantial area of healthy native shrub-steppe vegetation is also present. Thirteen study sites were sampled in 1997 on the basis of historical land use and twelve new sites were added in 1998. We used invertebrates collected in pitfall traps as our study group, investigating 56 characteristics of terrestrial invertebrate assemblages from taxa richness and composition to diverse functional roles. Our goal was to develop a terrestrial index of biological integrity composed of metrics, biological attributes that change quantitatively along a gradient of human influence. Of the 56 attributes tested, 28 distinguished between undisturbed and disturbed sites in 1997; no more than 7 would be expected by chance. Twenty attributes gave consistent responses to disturbance in 1998. Several taxa richness measures changed consistently along disturbance gradients: total number of invertebrate families, Diptera, Tachinidae (Diptera), Acarina, Tenebrionidae (Coleoptera), parasitoids, decomposers, and predators. Relative abundance of Eleodes (Tenebrionidae) decreased and dominance increased along the gradient. Integrating these measures in a multimetric index provides an effective measure of biological condition that can be used to guide or evaluate restoration and cleanup efforts at Hanford.


Multimetric indexes integrate multiple biological attributes (called metrics) to describe and evaluate the condition of a place. Metrics are chosen on the basis of whether they reflect specific and predictable responses of organisms to human activities. The first successful application of the multimetric concept to biological systems (index of biological integrity, or IBI) occurred in freshwater systems (Karr 1981; Karr et al. 1986). The IBI incorporates the responses of fish or benthic invertebrates to measure biological condition and is now used or being developed by most states and several federal agencies. Multimetric biological evaluations have stimulated a fundamental change in the way water resources are evaluated under the Clean Water Act.

Many recent studies have considered the potential of invertebrates as reliable indicators of disturbance or degradation in terrestrial systems (Holliday 1991; Niemela et al. 1993; Anderson 1997; Blair and Launer 1997; Rodriguez et al. 1998). Terrestrial invertebrates are prevalent, have high species diversity, are easy to sample, and are important in ecosystem function (Rosenberg et al. 1986). They respond to environmental changes more rapidly than vertebrates and can provide early detection of ecological changes (Kremen et al. 1993). They also have diverse roles in natural environments that include decomposers, predators, parasites, herbivores, and pollinators and these roles are affected by various perturbations.

Some studies have examined the responses of various taxa, including ground beetles (family Carabidae), ants (family Formicidae), and butterflies, to urbanization, logging, agricultural practices, and fire. Few, if any, have attempted to examine terrestrial invertebrate assemblages across a range of disturbance events. Our study is assessing the responses of a wide range of taxa collected at Hanford across a gradient of disturbed sites. The purpose is to select taxa or functional groups that respond in a consistent manner to various types of disturbance, and use those for developing a terrestrial index of biological integrity.

Study sites

This study adapted the IBI approach to terrestrial systems by focussing on insects at the DOE Hanford site in SE Washington state. The 560-square mile site, closed to public access and development since 1943, is in the "shrub-steppe vegetation zone." Presettlement vegetation consisted of shrubs (Artemisia spp., Chrysothamnus spp., Purshia tridentata) and perennial bunchgrasses (Agropyron spicatum, Festuca idahoensis, Stipa spp., Poa spp).

The first year of field sampling was conducted in Spring 1997, and another field season was completed in May 1998. Study sites were chosen in central Hanford to represent areas with a range of human-caused disturbance, including agriculture, heavy equipment, fire, chemicals, remedial activities, and grazing (Fig.1; Table 1). Thirteen sites were selected in 1997, and 19 sites in 1998. Site selection was based on discussions and field trips with many biologists familiar with Hanford, including individuals representing The Nature Conservancy of Washington (TNC), Washington State University (WSU), University of Washington (UW), Washington Native Plant Society (WNPS), and Battelle's Pacific Northwest National Laboratory (PNNL).

Field Methods: Insects

Four sampling methods were used for sampling terrestrial insects in 1997: (1) pitfall traps: three grids of 25 traps per site were set; (2) sweep-nets: 12 samples of 10 sweeps per site; (3) butterfly transects: four 10-m wide, 0.5-km transects were walked per site; and (4) individual shrubs: ten randomly selected Artemisia tridentata shrubs were examined per site. Galls were counted and a beating sheet was used to collect foliage insects within each quadrant (N,S,E,W) of the shrub.

Sampling methods were modified in 1998, based on results of the first season: (1) pitfall traps: three grids of 10 traps per site were set; (2) sweep-nets: 12 samples of 20 sweeps per site. Butterfly transects and individual shrub sampling were not repeated, as data from the first season did not add significant information to the data base.

The Order Hymenoptera (Bees, wasps, and ants) is of special interest because it has species that occupy a range of trophic levels, including pollinators, predators, and parasitoids. Two pollinators and a parasitoid species are represented above.

Table 1: Study site locations for 1997 and 1998 field seasons. The letter codes in column 1 correspond to the site locations on the map. Sites sampled for insects in 1997 and 1998 are marked with an x.

Preliminary Results: Terrestrial Insects

Samples sorted and identified in the laboratory from two field seasons have included 51,962 specimens (excluding springtails). Based on the collections, 570 species or morphospecies are represented from 127 families and 20 orders of insects, spiders, and other arthropods. These results are from 10 traps per grid per site, for a total of 30 traps per site.

The composition of insect assemblages showed many trends along a range of disturbance represented by study sites in 1997 and 1998. Twenty characteristics of insect assemblages showed consistent responses to disturbance in both years. Six examples are shown in the figures below. Taxa richness decreased in number of families (Figure 1), Tenebrionidae (darkling beetles) species (Figure 2), Diptera families (Figure 3), and parasitoids, including flies and wasps (Figure 4). Parasitoids involve third trophic level species interactions which may be more susceptible to disturbance than lower trophic level interactions. Dominance increased in disturbed sites (Figure 5). An example of this type of response is common in agricultural systems, where monocultures can favor specific "pest" species. The percentage of pollinating bee species was severely reduced at burn sites, but showed a favorable response to sites that were disturbed and had a high abundance of exotic annual flowering plants (Figure 6). All comparisons between disturbed and undisturbed sites were significantly different (Mann-Whitney p<0.05).

These and other patterns of biological change among the sites provide leads for identifying reliable metrics for use in a terrestrial IBI for Hanford. By choosing attributes that provide clear signals of the effects of human actions (dose-response curves), we can detect and understand the biological effects of diverse human actions.

Figure 1. Family richness declined with increasing levels of disturbance (UD=undisturbed, DIST=agricultural, physical disturbance, BURN=1984 burn sites, OTHER=unknown disturbance). Data shown are from 1997.

Figure 2. Tenebrionidae, darkling beetle, taxa richness was lower in disturbed and burned sites than in undisturbed sites (data shown from 1997). Darkling beetles are detrivores, eating a variety of plant materials.

Figure 3. There were fewer fly families in both disturbed and burned sites than in undisturbed sites. Flies have many roles in ecosystem function, including pollinators, parasitoids, decomposers, gall-makers, fungivores, and predators.

Figure 4. Parasitoid taxa richness, including parasitic flies and wasps, decreased in disturbed sites in 1997 and 1998. These insects represent a third trophic level interaction which may be more sensitive to food-web disruptions. (Data shown are from 1997.)

Figure 5. There was a strong trend for fewer families to dominate disturbed sites, although the difference between disturbed and undisturbed sites was not significant (Mann-Whitney p<0.10).

Figure 6. Pollinators responded differently to different disturbance events. Percentage of pollinators in the samples increased among sites used historically for agriculture and development, but was severely depressed where a 1984 fire occurred.

The photographs demonstrate the annual variation that occurs in insect sampling. The first set of petri dishes represent the contents of three pitfalls traps in 1997, taken from three sites. The first site is the Arid Lands Ecology Reserve (ALE), considered to be minimally disturbed. The second site (BM4) is a long-term study site that was burned in 1984. The third site (OF) is an old agricultural field that was abandoned in 1943 when Hanford was taken over by the DOE. The differences among the contents of the three dishes are obvious-the ALE sample is more diverse, the burn site is dominated by a single species of ground beetle, and the old field is dominated by a common agricultural pest, a cutworm. The second set of petri dishes represent the contents of three pitfall traps collected in April 1998 from the same sites. Differences among the sites are not as obvious; there were many population fluctuations, and the patterns discovered in the first year are not all consistent with the second year of data.


The results of this research provide a foundation for developing a terrestrial IBI.

Our study is the first effort to identify multiple attributes of invertebrate assemblages that respond to anthropogenic disturbance in a shrub-steppe landscape. We found 20 attributes (candidate metrics) that distinguished between disturbed and undisturbed sites in both 1997 and 1998. Most of the attributes, including taxa richness of various orders, families, and trophic levels, decreased with effects from human activities. Exceptions were an increase in dominance and in percent abundance of pollinators in disturbed sites. Responses to fire were analyzed separately because fire can be either natural or human-caused. We are continuing to assess attributes for metric selection to discover which ones provide the most consistent responses.

Our goal is to provide a strong scientific foundation for the assessment of ecological risk. The development of a terrestrial IBI will be instrumental in making DOE more responsive to the concerns of stakeholders and more effective at targeting expenditures for facility cleanup. Having an objective way to assess the biological condition of damaged sites is a key to successful restoration, mitigation, and conservation efforts.

Future Plans

We will continue to analyze data for metric selection and development of the terrestrial IBI. In addition, we are entering data from sweepnet samples. We plan to publish in peer-reviewed scientific journals and present our research at regional and national conferences. We also plan to test the metrics we develop from the Hanford study at other DOE facilities such as INEEL (Idaho National Engineering and Environmental Laboratory) in 1999.

We welcome insights that interested parties might provide concerning potential new study sites, other applications of multimetric biological monitoring and assessment, the ecological or cultural significance of resident plants and insects, and data or observations on historical land-use, fire, or other disturbance regimes.

Literature Cited

Anderson, A.N. 1997. Using ants as bioindicators: Multiscale issues in ant community ecology. Conservation Ecology (online) 1(1):art8.

Blair, R.B. and A.E. Launer. 1997. Butterfly diversity and human land use: Species assemblages along an urban gradient. Biological Conservation 80:113-125.

Holliday, N.J. 1991. Species responses of carabid beetles (Coleoptera:Carabidae) during post-fire regeneration of boreal forest. Canadian Entomologist 123:1369-1389.

Karr, J.R. 1981. Assessment of biotic integrity using fish communities. Fisheries 6(6):21-27

Karr, J.R., K.D. Fausch, P.L. Angermeier, P.R. Yant, and I.J. Schlosser. 1986. Assessing biological integrity in running waters: a method and its rationale. Illinois Natural History Survey Special Publication 5, Urbana Illinois.

Kremen, C., R.K. Colwell, T.L. Erwin, D.D. Murphy, R.F. Noss, and M.A. Sanjayan. 1993.Terrestrial arthropod assemblages: their use in conservation planning. Conservation Biology 7:796-808.

Niemala, J., D. Langor, and J.R. Spence. 1993. Effects of clear-cut harvesting on boreal ground-beetle assemblages (Coleoptera:Carabidae) in western Canada. Conservation Biology 7(3):551-561.

Rodriguez, J.P., D.L. Pearson, and R. Barrera R. 1998. A test for the adequacy of bioindicator taxa: Are tiger beetles (Coleoptera:Cicindelidae) appropriate indicators for monitoring the degradation of tropical forests in Venezuela? Biological Conservation 83(1):69-76.

Rosenberg, D.M., H.V. Danks, and D.M. Lehmkuhl. 1986. Importance of insects in environmental impact assessment. Environmental Management 10:773-783.

GO BACK TO Responsive Science