|
STAND MANAGEMENT COOPERATIVE
|
Back to NewsletterArticles of Interest
Biodiversity Indices: Roadmaps to Future Forests? JoEllen Kassebaum, Graduate Student, Evergreen College, Former SMC Field Crew MemberManaged forests have come under scrutiny with regards to decreased biodiversity (Pitkänen, 1998). Yet, there does not appear to be an established quantitative measurement system(s) to address this concern (Mazzotti & Morgenstern, 1997; Kangas & Pukkala, 1996; Halpern & Spies, 1995; Roberts & Gilliam, 1995; Underwood, 1995; Fairweather, 1993; Ehrlich & Ehrlich, 1992; Murphy & Noon, 1992; Hansen et al, 1991; Thomas & Salwasser, 1989). Recently, Stand Management Cooperative understory vegetative data collected from tree-spacing research plots were utilized to test existing biodiversity measures using a standard spreadsheet program. These data were collected from several sites in British Columbia, Washington, and Oregon using two different systems for species identification (life form characterization or botanical genus and species) and percent cover to estimate species abundance. Species identification (termed species richness) and abundance are integral parts of biodiversity calculations (Magurran, 1988; Pielou, 1975; Poole, 1974; Whittaker, 1972). Using the SMC-collected understory vegetation data from research sites with known sampling methods, eight alpha diversity measures (simple species richness and abundance counts, Margalef’s and Menhinick’s species richness indices, Shannon-Wiener index, Simpson index, Berger-Parker index, Q Statistics, and Hill’s order of indices system) were evaluated for 1) appropriateness to a conventional understory vegetation sampling method, 2) ease of calculation, 3) applicability to typical computer spreadsheet use, and 4) information produced. Comparing the Stand Management Cooperative sampling methods against established methods (see Gauch, 1989; Causton, 1988; Ludwig & Reynolds, 1988; Southwood, 1978; Mueller-Dombois & Ellenberg, 1974) indicates both the initial competition sampling grid and permanent survey plots used an establishment method similar to stratified random sampling. The environment, size and shape of each sample quadrant, and plant community structure under consideration are consistent. The field data collection methods are standardized and protocols are printed in the field manual available to the field crew each season. To this point, the sampling methods seem consistent with most ecological research procedures. The sampling methods need improvement in three areas:
Biodiversity calculations would be performed more easily within a statistical computer program. Most indices are applicable to typical spreadsheet programs, but the calculations require repetitions of multiplication and/or division. Tedious, but achievable. The same is true of squaring procedures within calculations. With larger biological data sets, the calculations require a computer with high memory capacity and a processor with some speed (i.e., greater than a 486). The entire Stand Management Cooperative database with 70,000 records does not translate to a normal-sized spreadsheet. Larger data sets are even more of a challenge. Additionally, the charting tool in the spreadsheet program (used to display some calculations) was limiting in the scale that could be chosen, the differentiation of data display bars, and the placement of titles. There may be more flexibility with a statistical program that is specifically written with statistical graphic displays in mind. Although it is possible to accomplish the diversity calculations on a typical spreadsheet program, a statistical software program or biodiversity program would reduce the time involved and the possibility of errors considerably. Prince (1986, p. 370) offers a table comparing the various software packages useful for ecological assessments. Biodiversity calculation packages are also available on the Internet. These diversity calculations produce unitless numbers that usually increase with increasing diversity. (Technically, they produce nats or nat bels if using a natural log in the calculation. However, use of any unit was not evident in the literature.) As such, there is no universal standard to compare against; a biodiversity index result needs to be compared to other results of the same index, either over time or space. Graphed over time they can indicate canopy closure, changes in habitat, management regime, etc. by the increases or decreases in their calculation number. Their use as a monitoring tool by comparing one year to the next might allow funds and personnel to be utilized more efficiently. A drastic decrease/increase in the index number for a site over a period could signal the need to investigate management practices, habitat changes, etc. Slight index variations might be interpreted as continued health and indicate that no immediate investigation is needed. Since the Stand Management Cooperative study is long-term, comparison of the research plots over time would produce characteristic data set patterns that could be correlated to specific stand health, which might allow development of a rating system similar to the Index of Biological Integrity used for streams (See Karr et al., 1986). For some resource managers, the possibility of assessing the fluctuation of plant communities with animal communities in the same area is of interest. Index evaluation might help establish correlation between management practices and animal populations and be useful with regard to Habitat Conservation Plans. Plant and animal communities could not be directly compared to one another, but if one shows an increase in diversity, it might be of interest to see if the other is also showing an increase. Using the same accepted diversity measurement system (e.g., Shannon-Wiener, Simpson, or species richness) for both populations would seem to reduce inconsistencies that might arise by using two different quantification systems. Another monitoring application might be tracking the fluctuation of the native understory community verses the invasive weed and/or noxious weed communities. High populations of one or more of these communities might be indicative of stand health, success of management regimes, or change in environmental conditions. Biodiversity measures are useful tools for resource management. They are not a definitive answer to diversity management but more one method to describe, monitor, or compare specific sites. Choice of diversity measure will be determined by the project goals or sampling methods. The Shannon-Wiener index is used in a variety of disciplines (communications, psychology, animal behavior study, engineering) but has some flaws when applied to ecological data (Magurran, 1988; Ludwig & Reynolds, 1988; Pielou, 1974; Poole, 1974). Specifically for the SMC data set, the limiting factor is the sampling methods under current practice. A well-trained survey crew with extensive botanical knowledge and a consistent sampling schedule would provide results that are more accurate. Quantifying biological functions or systems has the inherent problem of reducing complex relationships, interactions, or life processes into a more abstract concept (in this case, a number). Information is lost. Underlying mechanisms, health, or associations are not reflected. Although it seems more unbiased to assess sites quantitatively, resource management decisions involving biological entities should not be made solely on quantitative results. These indices are tools to describe, monitor, or compare a part of a whole system. As such, they can be indicators or guides to help discern areas or management regimes that need more thorough evaluation, suggest overall health or plant-animal relationships, reflect a general trend, or signal responses from silviculture and/or horticulture practices. They should not be interpreted as a definitive answer or proof. Although the Shannon-Wiener index appeared to be the most reflective of this data set and useful under a variety of other circumstances, there is no one perfect measure. Different diversity measures may be required at different times or for different purposes. As a potential monitoring device in resource management, a suggested general protocol might be:
Fairweather, P. G. 1993. Links between ecology and ecophilosophy, ethics and the requirements of environmental management. Australian Journal of Ecology, 18:3-19. Gauch, Jr., H. G. 1989. Multivariate Analysis in Community Ecology. Cambridge: Cambridge University Press. Goldsmith, F. B., C. M. Harrison, and A. J. Morton. 1986. Description and analysis of vegetation. Methods in Plant Ecology. 2nd ed. Moore, P. D. and S. B. Chapman, eds. Oxford: Blackwell Scientific Publications, 437-524. Halpern, C. H. and T. A. Spies. 1995. Plant Species Diversity in Natural and Managed Forests of the Pacific Northwest. Ecological Applications, 5(4):913-934. Hansen, A. J., T. A. Spies, F. J. Swanson, and J. L. Ohmann. 1991. Conserving Biodiversity in Managed Forests. BioScience, 41(6):382-392.
Back to SMC Newsletter |