A new method to estimate species and biodiversity intactness using empirically derived reference conditions

https://doi.org/10.1016/j.biocon.2007.02.024Get rights and content

Abstract

Critical to the conservation of biodiversity is knowledge of status and trends of species. To that end, monitoring programmes have reported on the state of biodiversity using reference conditions as comparison. Little consensus exists on how reference conditions are defined and how such information is used to index intactness. Most use protected areas or an arbitrary year as reference. This is problematic since protected areas are often spatially biased, while arbitrarily defined reference years are often not sufficiently distant in time. We propose an alternative that relies on empirical estimates of reference conditions. Statistical ranges of reference are estimated and compared with observed occurrence and abundance to index status of individual species. When averaged among species, overall intactness is estimated. We demonstrate the approach using 202-winter mammal tracking sites from the boreal forest of Alberta, Canada. Intactness was estimated at 89 out of 100 with the southern boreal having lowest intactness and greatest human footprint. We suggest empirical predictions of reference conditions be used as baselines for comparing changes in the state of species and biodiversity. Reporting can occur at any spatial (e.g., ecosystem) or hierarchical (e.g., species, guilds, taxonomic group, or overall biodiversity) scale and is easily interpreted (scaled from 0-degraded to 100-intact). When used in a long-term monitoring framework, statistical trends in biodiversity intactness can be estimated, individual status of species assessed, and relevant policy evaluated.

Introduction

Biodiversity is threatened by modern human activities (Hooper et al., 2005). The current extinction crisis is one of the most significant in earth’s history, with habitat loss, spread of non-native species, and global climate change the greatest threats (Wilcove et al., 1998, Chapin et al., 2000). Maintenance of biodiversity is important as its erosion will result in less stable ecosystems with reduced function (Naeem et al., 1994, Naeem et al., 1995, Tilman et al., 1996, Stachowicz et al., 1999). Reduced function and stability eventually lead to greater uncertainty in ecosystem services, including a number critical for human welfare (Costanza et al., 1997, Millennium Ecosystem Assessment, 2005). The value of such services is substantial, with global natural capital estimated at $33 trillion (US) per year in 1997; nearly double the global gross national product (Costanza et al., 1997). Given economic values and social-ethical concerns, governments, organizations, and scientists have attempted to quantify the ‘state’ of biodiversity by assessing status and trends, setting targets for mitigating biodiversity loss, and/or identifying hot spots for biodiversity protection (Dobson et al., 1997, Dobson et al., 2001, Myers et al., 2000, Weber et al., 2004, Scholes and Biggs, 2005). Despite the need for consistency in monitoring programmes, no single method of measuring or reporting biodiversity has emerged (Purvis and Hector, 2000).

When biodiversity is measured and reported, it is not always evident what benchmark to use for comparison and indexing (Allen et al., 2003). Three general approaches have been used: (1) desired goal or target; (2) time-zero; and (3) protected areas. In desired goal or target, expert opinion or social values determine reference (benchmark) conditions (Young et al., 2004). Floristic quality assessments, for instance, have been used to assess ecological integrity of the Midwestern USA (Herman et al., 1997, Taft et al., 1997) using prior assignment of coefficients of conservation for each species (Wilhelm and Masters, 1995). Such assignments are impractical when dealing with hundreds to thousands of species necessary to inform biodiversity and for taxonomic groups about which little knowledge exists. Moreover, additional quantitative information, such as relative abundance (density, percent cover, etc.), is not fully considered. As an alternative to desired states, time zero referencing has been suggested. Here, a point in time is selected (normally the start of the monitoring programme) to compare and index against current conditions. The Living Planet Index uses 1970 as a benchmark to report on the state of the planet’s ecosystems and species (Loh et al., 2005). Without a sufficiently distant past, time zero references fail to fully inform conservation-based boundaries for restoration and status assessments. Local areas within many ecosystems were already highly degraded in the year 1970. Furthermore, comparisons between monitoring programmes are compromised unless year of time zero and level of degradation are similar. Protected areas have also been used as comparison benchmarks. Sites of interest are compared against ‘natural’ or ‘intact’ reference sites, such as national parks (Mayer and Galatowitsch, 2001, Sinclair et al., 2002, Scholes and Biggs, 2005). Existing protected areas do not always contain a representative sample of biodiversity (Scott et al., 2001, Hansen and Rotella, 2002), since they often occur in remote high elevation areas lacking the potential for cultivation (Margules and Pressey, 2000, Scott et al., 2001). Without controlling for environmental gradients, differences among target and control areas can be solely due to natural patterns in species distributions, rather then anthropogenic influence. Furthermore, protected areas are being degraded over time by human activity resulting in sliding benchmarks.

We propose a fourth alternative for calculating benchmarks and biodiversity intactness. By estimating empirical relationships between species occurrence/abundance and human footprint we are able to estimate reference conditions under a pristine situation. These statistically-derived reference conditions are then compared to current species occurrence and abundance to index intactness. Deviation from reference (decreasing sensitive species or increasing non-native species) results in loss of intactness. With species as the basic unit of measure, numerous levels of organization can be reported (i.e., guilds, taxonomic group, or overall biodiversity). We demonstrate the utility of the approach using winter mammal monitoring data collected from the boreal forest of Alberta, Canada.

Section snippets

What to measure?

Biodiversity encompasses numerous levels of natural organization. Species, however, are the focus of biodiversity, because they are the most easily defined (Noss, 1990, Huston, 1994). Although other levels of biodiversity organization, such as genetic diversity (Watson-Jones et al., 2006) and landscape configuration (Roy and Tomar, 2000, Lindenmayer et al., 2006), are important, we focus the development of our biodiversity index on species. Measures of species occurrence and abundance, rather

Intactness for mammals in the boreal forest of Alberta: a working example

To illustrate our method, we demonstrate the biodiversity intactness index using winter mammal monitoring data in the boreal forest of northeastern Alberta, Canada.

Discussion

Indices of biodiversity integrity are desired by policy-makers as a mechanism to monitor change in ecological condition. Developing appropriate indices has proven to be difficult in practice (Purvis and Hector, 2000). There have been at least three significant challenges to characterizing biodiversity condition: (1) establishment of appropriate reference conditions, (2) sensitivity to both rarity and overabundance, and (3) incorporating both native and non-native species in a single measure of

Conclusions

As many existing monitoring programmes already collect information on species occurrence and abundance, we suggest occurrence and abundance be used as a foundation for monitoring biodiversity intactness. By estimating a range of reference for each measure based on empirical models, current conditions can be compared with the estimated range of reference to determine whether current conditions are “normal”. By scaling deviations between 0 (degraded) and 100 (intact), while leaving observations

Acknowledgements

Alberta Biodiversity Monitoring Program (ABMP) and the University of Alberta provided funding, making this work possible. We thank R. Noss and C. Aldridge for reviewing a draft of this manuscript and making helpful suggestions that improved the manuscript.

References (48)

  • R. Costanza et al.

    The value of the world’s ecosystem services and natural capital

    Nature

    (1997)
  • S.E. Cote et al.

    Ecological impacts of deer overabundance

    Annual Review of Ecology Evolution and Systematics

    (2004)
  • A.P. Dobson et al.

    Geographic distribution of endangered species in the United States

    Science

    (1997)
  • A.P. Dobson et al.

    Synoptic tinkering: Integrating strategies for large-scale conservation

    Ecological Applications

    (2001)
  • R.T.T. Forman et al.

    Road ecology: Science and solutions

    (2003)
  • M.E. Gompper

    Top carnivores in the suburbs? Ecological and conservation issues raised by colonization of north-eastern North America by coyotes

    Bioscience

    (2002)
  • M.J. Gundale

    Influence of exotic earthworms on the soil organic horizon and the rare fern Botrychium mormo

    Conservation Biology

    (2002)
  • A.J. Hansen et al.

    Biophysical factors, land use, and species viability in and around nature reserves

    Conservation Biology

    (2002)
  • K.D. Herman et al.

    Floristic quality assessment: Development and application in the state of Michigan (USA)

    Natural Areas Journal

    (1997)
  • D.U. Hooper et al.

    Effects of biodiversity on ecosystem functioning: A consensus of current knowledge

    Ecological Monographs

    (2005)
  • M.A. Huston

    Biological diversity: The coexistence of species on changing landscapes

    (1994)
  • H. Linden et al.

    Wildlife triangle scheme in Finland: methods and aims for monitoring wildlife populations

    Finnish Game Research

    (1996)
  • Loh, J., Randers, J., MacGillivray, A., Kapos, V., Jenkins, M., Groombridge, B., Cox, N., 1998 Living Planet Report...
  • J. Loh et al.

    The Living Planet Index: using species population time series to track trends in biodiversity

    Philosophical Transactions of the Royal Society B: Biological Sciences

    (2005)
  • Cited by (52)

    • Mask R-CNN based automated identification and extraction of oil well sites

      2022, International Journal of Applied Earth Observation and Geoinformation
    • Setting reference levels and limits for good ecological condition in terrestrial ecosystems – Insights from a case study based on the IBECA approach

      2020, Ecological Indicators
      Citation Excerpt :

      More specifically, using baseline years is less appropriate for monitoring progress towards management goals that are set in absolute terms (e.g. evaluating restoration success) and for comparisons across sites or geographical areas as conditions, and hence baseline values, will generally differ for any given baseline year (cf. Soga & Gaston 2018). For the purpose of general applications and/or comparative purposes, a quantification of a universally defined desirable ecosystem reference condition is therefore preferable (Scholes & Biggs 2005, Nielsen et al. 2007). Such universally-defined reference values may be developed to describe the system in good ecological condition as well as by associated limit values – or tipping points – beyond which the system is no longer considered to be in an acceptable condition (e.g. Becker & Hoffmann 2019).

    • What Is Invasion Biology?

      2018, Ecological Economics
      Citation Excerpt :

      Invasion biologists often refer to indices of biodiversity “intactness” or “integrity” which make the occurrence of introduced species a per se indicator of biodiversity decline (Gibbons and Freudenberger, 2006; Newbold et al., 2016). “Intact biodiversity is defined as the community found in the absence of anthropogenic disturbance” (Lamb et al., 2009: 442, citing Nielsen et al., 2007). If invasion biology excludes introduced or invasive species by stipulation from the idea of biodiversity or rules them out a priori from the species richness of a site, it relies on stipulative tautologies not on empirical discoveries to describe their “negative” effects.

    • Experts and models can agree on species sensitivity values for conservation assessments

      2018, Biological Conservation
      Citation Excerpt :

      Likewise, modelling is only as good as the input data and knowledge of the study system. There will always be cases in which neither method is accurate, such as rare or elusive species where experts lack the experience to make an informed decision and empirical results become imprecise (Nielsen et al., 2007; Matthews et al., 2015). However, it is patterns of excessive mismatch that signal deficiencies in expert knowledge, errors in the modelling process, or both.

    View all citing articles on Scopus
    1

    Tel.: +1 780 492 4165; fax: +1 780 492 9234.

    2

    Tel.: +1 780 632 8306; fax: +1 780 632 8379.

    3

    Tel.: +1 780 492 5766; fax: +1 780 492 9234.

    4

    Tel.: +1 780 492 1297; fax: +1 780 492 9234.

    View full text