Portal de Periódicos da CAPES

Indices: Attractive delusions

2011; Wiley; Volume: 7; Issue: 3 Linguagem: Inglês

10.1002/ieam.197

1551-3793

Peter M. Chapman,

Heavy metals in environment

The word “index” comes from Latin and means a pointer, sign, indicator, list, or register. It is used in economics, anthropology, physics, mathematics, astronomy, and a number of other disciplines. An index as used in science can be defined as a number, derived from a formula, that summarizes some quantity of data. For instance, a diversity index can be defined as a statistic that is intended to measure the differences among different biological communities. Indices are most commonly used for the purposes of making comparisons between 2 or more data sets over time. An index is, in the simplest terms, a ratio that measures change and allows us to compare the magnitude of different kinds of changes. Indices are attractive in science and especially for the communication of science to nontechnical audiences. Indices summarize complex data sets, usually into a single number that is easy to understand, particularly for judges in a court of law and members of the public. Alas, indices rarely reflect reality. For both regulated and regulating communities, indices are misleading at best and attractive delusions at worst. Indices should not be used as “bright lines” for establishing regulatory requirements such as discharge permits or environmental remediation objectives. At the very least, indices should not be used without detailed explanation to convey the important aspects of variability inherent in all environmental conditions. For instance, Washington (1984) provides an excellent review of diversity, biotic, and similarity indices, showing how biological indices are often misused because they are: 1) typically highly specialized to compare and contrast conditions associated with a particular type of water pollution, usually organic pollution; 2) limited to specific geographic areas; and 3) of limited ecological relevance because they do not clearly separate natural environmental change from change related to various types of anthropogenic stressors. Used outside their appropriate context, environmental indices may not accurately describe the single or multiple environmental conditions they are intended to communicate. They are, rather, merely attractive delusions. Unfortunately, Washington's (1984) review of biological indices and the use of indices in general was published well before the advent of Internet search engines and thus is generally forgotten by the scientific community. Green and Chapman (2011) resurrected the Washington article, together with even earlier and more recent publications in a review of the proper use of environmental indices, and concluded that credible scientists should not be developing or relying on single-number representations of complex data, and they should not be misleading nonscientists that this is appropriate or even useful. The reduction of complex information to an index by necessity results in the loss of information, which can and does result in misleading interpretations of the original, much more complex, data set. Using more than 1 index usually is not helpful, because they are often highly correlated, communicating nearly the same narrowly focused level of information (Green and Chapman 2011), which can be reassuring to users even though the information itself is less useful than desirable. Alas, yet another attractive delusion. There are no “perfect indices” any more than there are any perfect tools for assessing differences among different biological communities. In Europe, emerging regulatory initiatives such as the Water Framework Directive, Habitat Directive, and others encourage the development of simplistic indices describing conditions such as biological diversity, water quality, and good environmental status. In the case of biological diversity indices, diversity may be in the eyes of the beholder. Looking out the window, for instance, and viewing croplands, woodlands, wetlands, and grasslands, we might conclude the scene possesses a high biological diversity index. Viewed from another perspective, perhaps because of limited vision or obstructions to our view, we might look at the same scene and only see a partial landscape and conclude the biological diversity index is lower as compared to the unobstructed view. What is the likelihood in a multicultural setting such as the European Union that indices developed to describe good environmental status may be viewed differently among member states? Caution is warranted in the strongest possible terms against believing that ecological complexity can be adequately summarized when indices reduce large volumes of seemingly disparate environmental data to single numbers. If business and regulatory agencies must develop or rely on environmental indices, it is important to convey their uses and limitations. In other words, users of indices must understand and explain the additional, and far from insignificant, uncertainties associated with simple summaries of complex data sets. Do not perpetuate attractive delusions, but instead, look deeper at the multiple lines of information that, when drawn together, convey a more complete picture of the complexities and interconnectedness inherent in the environment. Remember how at the end of the 1939 motion picture The Wizard of Oz, Dorothy's little dog Toto pulled back a curtain, exposing the dreaded wizard as an ordinary human trickster? Indices are often the trickster, hiding the true environmental quality behind a curtain. And, as noted by Green and Chapman (2011), if you really must use indices, do not use them in isolation; use them together with other statistical methods that retain more of the information in the original data set.