SEARCH

SEARCH BY CITATION

BIOANALYTICAL TOOLS IN WATER QUALITY ASSESSMENT

  1. Top of page
  2. BIOANALYTICAL TOOLS IN WATER QUALITY ASSESSMENT
  3. FISH PHYSIOLOGY: HOMEOSTASIS AND TOXICOLOGY OF ESSENTIAL METALS, VOLUME 31A FISH PHYSIOLOGY: HOMEOSTASIS AND TOXICOLOGY OF NON-ESSENTIAL METALS, VOLUME 31B
  4. STRUCTURED DECISION MAKING: A PRACTICAL GUIDE TO ENVIRONMENTAL MANAGEMENT CHOICES
  5. PRICELESS: ON KNOWING THE PRICE OF EVERYTHING AND THE VALUE OF NOTHING
  6. MOSTLY HARMLESS ECONOMETRICS: AN EMPIRICIST'S COMPANION
  7. REFERENCES

by Beate Escher and FredericLeusch

Thumbnail image of

This is a textbook for an approach to environmental assessment that is not yet generally recognized or practiced. Escher and Leusch's bioanalytical tools are “cell-based and low-complexity in vitro bioassays indicative of specific endpoints relevant to human and/or environmental health.” They argue that the number of chemicals in the environment and in drinking water and the complexity of mixtures make the usual approach of analytical chemistry and whole organism tests of individual compounds ineffective. Rather, they define a set of modes of action of organic chemicals and then propose the use of test batteries to determine whether a water sample induces effects by one or more of those modes of action. The results might be sufficient to support a regulatory action or, more likely, would lead to a conventional risk assessment that is focused on the identified mode of action and the chemicals that may induce it. This approach is, as the authors discuss, conceptually related to whole effluent toxicity testing and toxicity identification evaluation, but is potentially more efficient.

I called this a textbook because it goes beyond presenting the authors' approach to provide the supporting concepts that are needed to make a full assessment practice. These include exposure-response modeling, analysis of combined toxic effects, quality assurance and quality control, and toxicokinetic and toxicodynamics pathways. These chapters are clear and well-illustrated and could serve as stand-alone tutorials.

The concluding chapter confronts the difficult issues that lie between the bioanalytical approach and its general application. These include technical problems with the tests (e.g., matrix effects), the need to develop new tests, problems in linking cellular responses to both environmental chemistry and whole organism toxicology, and linkage to regulatory practices. It seems clear to me that the next step, if this approach is to be developed into a regulatory practice, is for a regulatory agency to decide to make it happen. Scattered efforts in academic laboratories and research institutes will not get us there.

2012. 253 pp. Soft Cover. ISBN 978-1-84339-368-9. $171. IWA Publishing, London, UK.

Glenn Suter

SETAC Reviews Editor

FISH PHYSIOLOGY: HOMEOSTASIS AND TOXICOLOGY OF ESSENTIAL METALS, VOLUME 31A FISH PHYSIOLOGY: HOMEOSTASIS AND TOXICOLOGY OF NON-ESSENTIAL METALS, VOLUME 31B

  1. Top of page
  2. BIOANALYTICAL TOOLS IN WATER QUALITY ASSESSMENT
  3. FISH PHYSIOLOGY: HOMEOSTASIS AND TOXICOLOGY OF ESSENTIAL METALS, VOLUME 31A FISH PHYSIOLOGY: HOMEOSTASIS AND TOXICOLOGY OF NON-ESSENTIAL METALS, VOLUME 31B
  4. STRUCTURED DECISION MAKING: A PRACTICAL GUIDE TO ENVIRONMENTAL MANAGEMENT CHOICES
  5. PRICELESS: ON KNOWING THE PRICE OF EVERYTHING AND THE VALUE OF NOTHING
  6. MOSTLY HARMLESS ECONOMETRICS: AN EMPIRICIST'S COMPANION
  7. REFERENCES

edited by Chris M.Wood, Anthony P. Farrell, and Colin J. Brauner

Thumbnail image of
Thumbnail image of

The dance between nutritional essentiality, mimicry, and toxicity of metals began to be recognized by at least the 1930s. The cells of humans and all organisms require that major essential metals such as Ca and Na be closely regulated, as well as trace metals such as Cu and Zn. Maintaining this mineral balance is challenging enough for mammals with physicians who sometimes supplement matters with Ca channel blockers or other pharmaceuticals. This mineral balance becomes particularly remarkable with fish that must constantly regulate their mineral balance while living in too-dilute freshwater solution or too-concentrated saltwater solutions. The central theme of this book is the interplay of metals that are essential to metabolic reactions or enzymes (but that can also be disruptive and deadly when in excess) and nonessential metals such as Cd and Hg.

This “book” (actually 2 companion volumes) provides a comprehensive and accessible review of trace metal essentiality, effects of deficiency or excess, homeostatic processes, and toxicology in fishes. The chapters and volumes are constructed with a parallel structure that helps comparisons across the different metals. In addition to the core focus, each chapter includes a brief summary of geochemical speciation, environmental concentrations in natural and polluted areas, environmental quality criteria from different countries, uses, and arguments for and against essentiality. The chapters are all authoritative.

Sixteen of the 70 metals in the periodic table are addressed. In addition to the usual suspects, less well known metals (from an aquatic toxicology perspective) such as Fe, Co, U, Sr, and Mo rated consideration. I particularly appreciated the 3 synthesis chapters: a Basic Principles introduction by Chris Wood; Field Studies by Patrice Couture and Greg Pyle, and Modeling the Physiology and Toxicology of Metals by Paul Paquin and colleagues. Each of these 3 chapters would make good outlines for upper level undergraduate or graduate courses in environmental science programs, as well as Society of Environmental Toxicology and Chemistry (SETAC) continuing education primers.

I recommend this book without qualification to anyone with more than a fleeting interest in the effects of metals on fish. I did have a few quibbles with instances where the studies described were disconnected from environmental relevance. For instance, in predatory or long-lived freshwater fish, Hg tissue residues resulting from trophic transfer of ambient water column Hg concentrations in the low ng/L range have been linked to neurotoxicity, maladaptive behaviors, and reproductive impairment. Yet biomagnification was given short shrift with a half page, in contrast to a 5-page compilation of ecologically meaningless acute toxicity values, which were 4 to 6 orders of magnitude higher than environmentally relevant water concentrations. A reversal of this emphasis would have been welcomed. Similarly, after reading that Ni concentrations of concern are usually in the tens to low hundreds of µg/L range or less, I was not sure what to make of research on respiratory toxicity in fish conducted at Ni concentrations greater than 10 mg/L. Such sledgehammer studies do indeed produce reportable effects (p < 0.05), and perhaps such disconnects might in part be because of the focus on fish. Other aquatic groups such as zooplankton, amphipods, or mollusks are often more sensitive to metals in water than are fish.

The publication format of this ‘book’ is a dual hybrid between the conventional, academic book and special issue articles in the periodical “Fish Physiology.” For instance, the Cu chapter can be accessed online as a journal article (Grosell, M. 2011. Copper. Fish Physiology: 31(PART A): 53-133. Available from: http://dx.doi.org/10.1016/S1546-5098(11)31002-3) in addition to appearing as Chapter 2 in the hardcover version. This hybrid book/periodical publication format has a number of advantages over traditional book publishing by improving visibility, access, and citations via bibliographic search services such as Scopus, Web of Science, or CrossRef, plus the ability to request individual chapter/article reprints. I cannot help but think that SETAC Books might do well to consider moving to a similar model because our society's Pellston series and other books have a common theme along the lines of “Environmental Toxicology, Risk Assessment, and Management.”

These 2 volumes are likely to stand for some time as the defining compendium on the homeostasis and toxicology of metals in fish. The publisher lists them as the First Edition. Perhaps when the Second Edition is written, it will be feasible to expand the scope to include comparative information on aquatic organisms other than fish.

2012. 520 pp. (Volume 31A). eBook or Hardcover. ISBN 978-0123786364. Approximately $110. 5 Academic Press, Waltham, MA.

2012. 530 pp. (Volume 31B). eBook or Hardcover. ISBN 978-0123786340. Approximately $110. Academic Press, Waltham, MA.

Christopher Mebane

Boise, Idaho

STRUCTURED DECISION MAKING: A PRACTICAL GUIDE TO ENVIRONMENTAL MANAGEMENT CHOICES

  1. Top of page
  2. BIOANALYTICAL TOOLS IN WATER QUALITY ASSESSMENT
  3. FISH PHYSIOLOGY: HOMEOSTASIS AND TOXICOLOGY OF ESSENTIAL METALS, VOLUME 31A FISH PHYSIOLOGY: HOMEOSTASIS AND TOXICOLOGY OF NON-ESSENTIAL METALS, VOLUME 31B
  4. STRUCTURED DECISION MAKING: A PRACTICAL GUIDE TO ENVIRONMENTAL MANAGEMENT CHOICES
  5. PRICELESS: ON KNOWING THE PRICE OF EVERYTHING AND THE VALUE OF NOTHING
  6. MOSTLY HARMLESS ECONOMETRICS: AN EMPIRICIST'S COMPANION
  7. REFERENCES

by Robin Gregory, Lee Failing, Michael Harstone, Graham Long,Tim McDaniels, and DanOhlson

Thumbnail image of

Once upon a time, risk assessors confidently used the phrase “risk-based decision making.” Now we know that decisions are, at best, risk informed. Cost-benefit analysis can claim to be a basis for decision making, but it is used more as a means to block decisions. In fact risks, costs, benefits, and other analytical products from predictive assessments are only a part of the input to decision making. More mundane inputs include political commitments and ideologies, stakeholder preferences, laws, regulations, legal precedents, and public opinions.

The seemingly obvious solution is to use multicriteria decision analysis (MCDA). However, MCDA was developed to facilitate choices by a decision maker confronted by a choice involving multiple criteria. It has been used for cases involving multiple contributors to a decision as well as multiple criteria (see Linkov and Moberg 2012, which is reviewed in the previous issue of this journal, IEAM Vol 8, no. 2). However, that is a forced fit. Gregory et al. have developed a method to deal with such problems that they call Structured Decision Making. It incorporates MCDA, risk analysis and other types of analytical assessment, but its core is social psychology.

As the name implies, the primary task of structured decision making is to elucidate the structure of the environmental problem in a way that allows all goals and constraints to be clear to each participant. The process does not aim for compromise because that often leads to vaguely defined or relatively safe solutions rather than creativity. In fact, although making trade-offs is an important part of the process, the method can be a success even if there is no agreed-upon solution. The ultimate decision maker will not necessarily agree with the decision of a group of stakeholders. However, stakeholder input from this process would be much more useful than unstructured input from individual stakeholders and assessment teams.

I recommend this book to anyone who must function at the interface between environmental science and decision making. Even if you do not have the opportunity to implement the full structured decision-making process, it will give you a better idea of how to think about your role and those of the other parties. The authors write clearly and forcefully. For example, “Science alone will not make good environmental policy choices. But a values free-for-all will not get us there either.”

2012. 299 pp. Soft cover. ISBN 978-1-4443-3342-8. $70. Wiley-Blackwell, Chichester, UK.

Glenn Suter

SETAC Reviews Editor

PRICELESS: ON KNOWING THE PRICE OF EVERYTHING AND THE VALUE OF NOTHING

  1. Top of page
  2. BIOANALYTICAL TOOLS IN WATER QUALITY ASSESSMENT
  3. FISH PHYSIOLOGY: HOMEOSTASIS AND TOXICOLOGY OF ESSENTIAL METALS, VOLUME 31A FISH PHYSIOLOGY: HOMEOSTASIS AND TOXICOLOGY OF NON-ESSENTIAL METALS, VOLUME 31B
  4. STRUCTURED DECISION MAKING: A PRACTICAL GUIDE TO ENVIRONMENTAL MANAGEMENT CHOICES
  5. PRICELESS: ON KNOWING THE PRICE OF EVERYTHING AND THE VALUE OF NOTHING
  6. MOSTLY HARMLESS ECONOMETRICS: AN EMPIRICIST'S COMPANION
  7. REFERENCES

by Frank Ackerman and Lisa Heinzerling

Thumbnail image of

Increasingly, environmental decision making is based on cost–benefit analysis. Environmental laws passed in the 1970s in the United States mandated protection of human health and the environment, but, beginning in the Reagan administration, a series of executive orders have mandated benefit–cost analysis as the standard for acceptability of major environmental decisions. That mandate is assigned to the Office of Management and Budget (OMB). The authors of this book (an economist and an attorney) critically review the history of cost–benefit analysis of environmental decisions in the United States.

The first problem is the complete accounting of the costs of a regulation or standard to those who must comply and the incomplete accounting of benefits to human health and welfare and the environment. An example is the OMB's 2002 review of the Forest Service's decision to protect certain roadless areas. They concluded that the cost would be $184 million and the benefits would be only $219 000. So, how could the protection of 60 million acres of forests, watersheds, plants, and animals provide so little benefit? The answer is that the only economic benefit was the avoided cost of building roads to develop the roadless areas. Other examples are less blatant, but it seems inevitable that the monetary benefits are a subset of the full benefits of environmental protection. (See the article by Bob Costanza in IEAM Vol. 2, no. 2.)

Another problem is the distribution of costs and benefits. The costs of environmental protection accrue to 1 group and the benefits to another that is often those with lower income (inhabitants of inner cities with high Pb levels, subsistence fishermen, etc.) or with no income (trees, birds, etc.). The goal of economic efficiency is met if wealthy people gain $2 per $1.99 lost by poor people. This is Pareto optimal because the winner could, in theory, pay off the loser and still come out ahead. However, of course, they do not make such payments. Economists argue further that economic efficiency is a health goal because more money means better health. However, they neglect to mention that more income significantly increases health only until you reach a middle class income. Making the rich richer does not increase health.

A third problem is the poor quality of the science that economists use in cost–benefit analyses. Some of this comes from bias by the economists who perform the studies. For example, the authors describe how Tengs and Graham reported that government regulations cost $99 billion for every year of life saved. They obtained that number by including in their compilation regulations that were proposed but withdrawn and even regulations that were considered but never proposed, exactly because they would have been so expensive. Tengs and Graham also include regulations that were intended to prevent nonlethal health effects (and therefore had no benefits by their standard) and they discounted future deaths. (As Ackerman and Heinzerling point out, people cannot be banked and they do not accrue interest, so applying the discount rate for money to people is nonsensical. However, it is standard practice.)

Even the less ideologically biased methods would not be considered good science outside of economics. A standard means of estimating the value of human life is wage risk analysis that determines how much more is paid for more dangerous jobs. It is based on the implausible assumption that workers have a free choice of jobs at a range of risk levels and salaries, and that they are fully aware of the probability of death for each. It also does not include those who would not accept a dangerous job at any of the wages offered. For example, nobody would pay me enough to be a roofer because walking round on a steep slope high above the ground seems too dangerous to me.

Because many benefits have no market, they are often estimated by simply surveying people about their willingness to pay. The problems with this approach are obvious. One environmental economist explained it to me this way. If you were asked how much you would be willing to pay to avoid a one in a million risk of cancer, you would probably provide a number for that hypothetical. However, if I showed up at your house in a truck and asked how much you would be willing to pay me to not spray your house with a carcinogenic chemical, you would be more likely to call the police than pay me anything. However, this approach has been embraced by environmental economists because it is the only accepted way to estimate most environmental benefits.

The authors present other objections and many clever and illustrative examples. In the end, they conclude that decisions concerning the use of the environment are political, not economic, decisions. In particular, “the expression of social priorities through existence values asks us to view ourselves only as consumers; it attempts to replace voting with shopping.”

A rebuttal is provided by a review in The New Republic by Cass Sunstein, who has replaced John Graham as the regulatory czar at OMB. Sunstein concedes many of Ackmean and Heinzerling's points. However, his conclusion, crudely summarized, is that conventional economics has won and there is no point in fighting. Rather, we should work to make cost–benefit analysis better. I am afraid that he is right. Research at the US Environmental Protection Agency and elsewhere is trying to improve the estimation of ecological benefits through the definition and quantification of ecosystem services. Many of us support those efforts, but wish we did not have to.

2004. 277 pp. Soft Cover. ISBN 978-1-56584-981-5. $18. The New Press, New York, NY.

Glenn Suter

SETAC Reviews Editor

MOSTLY HARMLESS ECONOMETRICS: AN EMPIRICIST'S COMPANION

  1. Top of page
  2. BIOANALYTICAL TOOLS IN WATER QUALITY ASSESSMENT
  3. FISH PHYSIOLOGY: HOMEOSTASIS AND TOXICOLOGY OF ESSENTIAL METALS, VOLUME 31A FISH PHYSIOLOGY: HOMEOSTASIS AND TOXICOLOGY OF NON-ESSENTIAL METALS, VOLUME 31B
  4. STRUCTURED DECISION MAKING: A PRACTICAL GUIDE TO ENVIRONMENTAL MANAGEMENT CHOICES
  5. PRICELESS: ON KNOWING THE PRICE OF EVERYTHING AND THE VALUE OF NOTHING
  6. MOSTLY HARMLESS ECONOMETRICS: AN EMPIRICIST'S COMPANION
  7. REFERENCES

by Joshua D. Angrist and Jörn-Steffen Pischke

Thumbnail image of

Scientists who analyze data as a justification for environmental findings may find it useful to compare how data problems are addressed in various disciplines. Such efforts may identify useful new methods, deepen our insights into use of standard methods, or provide exercises in thinking through tricky inferences.

Sometimes, it will be helpful to have a credible estimate of a causal effect. For example, in epidemiology we may like to estimate of the increase in risk of some disease, associated with an increased exposure to a chemical. (Rothman et al., 2008, may be consulted for methods of quantitative epidemiology.) One hopes that the effort to obtain such an estimate does not expose one to an accusation of reliance on “mere correlation without causation.” Customary laboratory procedures of control and randomization are often considered critical for identifying causal relationships. Often, however, nonexperimental data (e.g., much ecological field data) will be used in some way. Recent interdisciplinary literature on methods of “causal inference” is of interest to those who hope to support causal conclusions using such data. Angrist and Pischke provide one useful contribution, focusing specifically on attempts to quantify effects of government policies and social background variables on measures of public welfare (e.g., employment effects of minimum wage laws). An apparent aim is to provide supplemental readings for graduate students in economics; however, the authors note that concerns of an applied econometrician will often be similar to those of an epidemiologist (we might add, “or environmental data analyst”).

I have found the literature of causal inference helpful especially with respect to possible causal interpretations of multiple regression results, especially as applied to field data on ecological stressors. In this book, multiple regression in particular is given a helpful treatment, including discussion of the effects of leaving out regressors that should be included, or of including regressors that are affected by a causal variable of interest. Additional methods and issues treated include matching (e.g., based on propensity scores), quantile regression, and variations on the bootstrap. Methods with a distinctively causal purpose include differences-of-differences and regression–discontinuity methods. The book contains numerous data analysis examples. Often, multiple regression is compared to more specialized methods, based on application to the same data.

A methodology not emphasized, but sometimes considered particularly relevant for causal modeling in ecology, is structural equations modeling (SEM) (Shipley, 2002; Grace, 2005).

Confounding is perhaps the most important threat to the accuracy of causal inferences from nonexperimental data. Statistical adjustments are often applied to account for confounding variables that have been recognized as important and measured. In practice, there may be some chance in a given situation that important confounding variables have not been measured, or even recognized as important. Some specialized methods have also been proposed for addressing the possibility of “unmeasured confounders.” One can fairly say that some approaches to unmeasured confounders involve a substitution of assumptions (an assumption of no unmeasured confounders may be replaced by other assumptions). Such a substitution is clearly involved, in particular, for one methodology especially popular among economists (and to some degree epidemiologists) for addressing unmeasured confounders, that of instrumental variables (IVs). The authors have contributed importantly to IV methodology in their previous research. The treatment in this book is extensive and potentially valuable to some analysts with a specific interest, but rather lengthy for purposes of an introduction.

The book is not very mathematical but some technical arguments have to be worked through. The humor, historical asides, and general “lack of gravitas” (as the authors put it) will be appreciated. Prerequisites such as multiple regression analysis will be for the most part familiar. The basic probability tools developed, in case they are not already familiar to the reader, are likely to be useful in many situations.

The book seems to complement, without markedly surpassing, some other books with a similar social sciences emphasis and level of technical difficulty. Most of the methods treated are also treated in Gelman and Hill (2007). A few practical issues are approached in different ways in the 2 texts, including use of survey weights, accounting for clustered variation, role of Bayesian modeling, and choice of statistical software. Morgan and Winship (2007) provide a somewhat broader treatment of causal inference, in particular discussing graphical methods that are not covered in this text (see also Shipley, 2002).

Two further qualifications seem appropriate. First, important findings regarding causal relationships among environmental variables are likely to be based on a largely qualitative process of weighing and integration of lines of evidence, a type of process that is not treated in this or other texts consulted for this review (that emphasize statistical analysis). Laboratory studies and biological insights may contribute to some lines of evidence. (The burden of evidence may not be born entirely by nonexperimental data.) Second, the methods described seem applicable primarily to large datasets. For decisions relating to circumscribed areas (“sites”), use of the methods may be comparatively limited. In the site assessment context, such methods may provide relatively indirect evidence, for example, information on effects to be expected based on patterns seen in monitoring data from the same geographic region as the site of immediate interest.

2010. 372 pp. Paperback. ISBN 978-0-691-12035-5. $37.50. Princeton University Press, Princeton, NJ.

David Farrar

National Center for Environmental Assessment Office of Research and Development US Environmental Protection Agency

REFERENCES

  1. Top of page
  2. BIOANALYTICAL TOOLS IN WATER QUALITY ASSESSMENT
  3. FISH PHYSIOLOGY: HOMEOSTASIS AND TOXICOLOGY OF ESSENTIAL METALS, VOLUME 31A FISH PHYSIOLOGY: HOMEOSTASIS AND TOXICOLOGY OF NON-ESSENTIAL METALS, VOLUME 31B
  4. STRUCTURED DECISION MAKING: A PRACTICAL GUIDE TO ENVIRONMENTAL MANAGEMENT CHOICES
  5. PRICELESS: ON KNOWING THE PRICE OF EVERYTHING AND THE VALUE OF NOTHING
  6. MOSTLY HARMLESS ECONOMETRICS: AN EMPIRICIST'S COMPANION
  7. REFERENCES
  • Gelman A, Hill J. 2007. Data analysis using regression and multilevel/hierarchical models. New York (NY): Cambridge University Press. 648 p.
  • Grace JB. 2006. Structural equations modeling and natural systems. Cambridge UK: Cambridge University Press. 378 p.
  • Morgan SL, Winship C. 2007. Counterfactuals and causal inference. New York (NY): Cambridge University Press. 328 p.
  • Rothman KJ, Greenland S, Lash TL. 2008. Modern epidemiology. 3rd ed. Philadelphia (PA): Lippincott Williams & Wilkins. 851 p.
  • Shipley B. 2002. Cause and correlation in biology. Princeton (NJ): Princeton University Press. 332 p.