Unreliable Eco Forecasts Produce Unnecessary Fear

During this latest period of heightened environmental interest given by a new round of international conferences, hoped for treaties and now also a papal encyclical, there is a rather curious embrace of the predictions of environmental science as forecasts that are to be believed, virtually at face value. “Studies show…”, “predictions indicate…”, “the established science reveals…” quickly place one in a defensive position. Who is in a position to question or reject this science? In the face of a constant call to religiously accept projections and expected outcomes, the only choice seems to be a believing embrace, a clasping of a form of environmental ideology.

This is a curious embrace since our culture is so demanding of evidence and rigorous justification of almost everything else. In fact, the current crisis of faith in a Creator God or in God as Savior or in a God as life-giving Spirit results heavily from a widespread inability to accept what one cannot see or demonstrate through human reason and understanding. Yet we seem to readily accept a wide range of environmental predictions.

We normally base our acceptance on verification of previous predictions; hence, our measure of at least some trust in daily weather predictions. Yet upon reflection, this embrace is not so curious since it is a response out of fear of the consequences of projections that are often perceived as dire. But predictive environmental science is not a kind of laboratory science that permits an acceptable verification of scientific hypotheses. Rather the science of environmental prediction has a nature that is inherently weak and as such is easily prone to misuse and inadvertent faulty representation. Why is this so? What is it that sets environmental science and its associated predictions apart by its very nature?

The management of environmental resources is ultimately given by a framework that seeks to have the maximum difference between the benefits from some proposed action and the costs associated with its control. This objective is deceptively simplistic in its obviousness. Of course, we want to have the benefits, that is, the positive results of our actions outweigh the costs, that is, the negative effects of an environmental control program. More than that, instinctively, we would want the benefits to not just be bigger than the costs but to be maximally better than the costs. But what is central to this desire is that the entire setting is predictive in nature; that is, the “benefits” are what we anticipate, expect, hope for and predict to occur if we in fact complete the environmental project. The “costs” are what we anticipate to be borne by an industry, a city or the population as a result of the action.

In both cases, the benefits and the costs of necessity need to be estimated or predicted simply because the control program has not yet been carried out to completion. In order to provide some basis for assessing the effectiveness or non-effectiveness of differing actions, analyses of the natural and social consequences are conducted using computer models of greatly varying levels of complexity. These models need to be calibrated and verified by comparing forecasts to observed conditions. There are however some key principles that should be followed.

Principles of Sound Environmental Science
We can identify four principles underlying any science that attempts to predict accurate outcomes. The science should be:

Believable: an analysis framework that is defensible through natural and social theory; such defensibility being verified and attested to by relevant scholars and practitioners,

Testable: reproducible in an experimental sense; that is, changes can be made in key factors, results measured and compared to theory and then corroborated by a repeat of the experiment,

Observable; measurements can be made of the present and future environmental reality,

Verifiable: after some control action is taken, observations can be made to test the veracity of the predictions.

For environmental science aimed at predictions, these principles quickly become problematic. The believability of the structure of the natural environment (e.g., the movement, transfer and transformation of a chemical in a river or lake) may be defensible and acceptable by the appropriate scientific establishment. The framework for some aspects of the social environment (e.g., the health impact of consuming fish with certain chemical concentrations) may also be recognized as a valid predictive model. But as the problem context increases in space and time, as well as extensions from the natural to the socio-economic environments, the structure itself—because of its growing complexity—becomes subject to increasing challenges. Climate change offers an illustration where the links between greenhouse gas emissions, climate alteration and impact on the health of poorer nations are hardly universally accepted despite claims to the contrary.

The testability of environmental frameworks is seriously restricted. The “laboratory” is the natural and social environment, local, regional and global with all of its complexities, interactions, ecologies and nested ecosystems. Depending on the scale and the time frame, the ability to hold certain variables in control while measurements are made is virtually non-existent. Laboratory experiments (e.g., health impacts in animals from chemical exposure) or social science testing can provide the components for building an overall framework but once built, environmental testability in an experimental laboratory sense is generally not possible.

For the third principle, that of the observability of the present environmental conditions, some substantial difficulties are again encountered. This “present” is essential since it provides the benchmark for evaluations of past and future changes. But, the grasping of the “present” is an elusive endeavor. Is one “snapshot” of an environment with a series of albeit limited measurements of the physiochemical, biological and social components sufficient to characterize the “present”? Or are multiple snapshots, a kind of “video” of the environment over a period of the “present” required to properly capture the ongoing ever-changing reality?

The fourth principle of verifiability is also difficult to be met except for small space and time scales. To verify is to determine the veracity or “truth” of what is claimed. Verifiability is linked to the predictability of the overall scheme. The science framework, natural or social, clearly lays some claim of predictability or else it would be of little value in the environmental arena. A framework that only explains what has been observed but could not be applicable on a larger scale would only have limited academic interest.

Thus, if a chemical or other waste is discharged into a small stream, the effects may be clearly seen, observable and capable of being controlled. With a prediction made as to the effectiveness of external control, the accuracy of the prediction is possible through direct measurement of the ecosystem. But even here one may have to wait some years, even decades to assess the accuracy of the prediction since it may depend, for example, on the occurrence of a low flow season that may occur only once every 10 or 20 years. As the space and time frames increase to regional or global scales over decadal time frames, the ability to verify predictions becomes virtually non-existent.

The degree of confidence in the prediction therefore must rely on these principles: believable, testable, observable and verifiable. The structure may be quite believable as attested to by multiple authors with the highest credentials, which provides the first level of confidence in the predicted outcome. But attempts at estimating a reasonable level of predictive confidence on decadal and global scales will almost certainly stumble on the latter three principles.

Fear of the Environment
So what? What difference does it make if we are not totally accurate in our predictions of environmental outcomes? Since “science” agrees in some way that the problem is significant and even if we are incapable of confidently predicting the outcomes, we cannot afford to delay. If I am afraid of what I don’t actually see or feel and of what might happen in the future, then why not embrace the predictions, just to be on the safe side? But it is the consequences of our actions that matter in the end and the consequences of embracing weak and fragile environmental projections and presenting such predictions with an air of certainty are serious indeed.

The claims introduce a level of environmental fear disproportionate to the projected effects. As fear increases, love declines and the environmental quickly devolves into a less than Christian state of affairs. No one wants to become more impaired intellectually due to chemicals in the air or water, nor does one want to lose income therefrom and certainly no one wants to have a heart or asthma attack or to finally die from chemical exposure. Massive extinctions of plant and animal species and continued environmental deterioration are all to be feared. These are primordial fears. Such fears are directly related to our level of confidence in predictions.

It is this deepening fear that the earth is at a “tipping point,” this fear of individual and societal catastrophes that result from the unreliable nature of the science of predictive environmental consequences. Claims have been made (EPA 2010) of significant loss of human life (4700 heart attacks and 11,000 premature deaths per year), unless standards for control of air toxics are instituted. It is beyond scientific credibility and believability to also offer a projected prevention of the loss of 0.00209 IQ points per child. At an IQ of 100, this amounts to claiming a change of 0.002 percent which would not be possible to measure when the numerous surrounding variables (education, diet, mental health) are recognized.

Such an ideological climate of fear readily distracts our attention from environmental problems of such significant magnitude that the four aforementioned principles are easily met. For example, we have known for many decades of the significant reality of communicable water borne diseases, especially in poorer countries and the consequences of such environmental problems on individual and societal health. Indeed, a significant part of the world’s population lives without adequate access to safe drinking water or sanitation facilities. Thousands die every day from this lack of safe drinking water and associated poor hygiene. These data are current environmental realities not forecasts. Further, we have a long record of success in reversing water borne diseases with proper controls. But the numbers are so numbing and the world has become so used to them that the impact of this environmental problem is largely dulled.

So if environmental predictions are embraced with little awareness of their weakness and inherently faulty nature, a toxic mix of fear and hopelessness will result. Predictions are essential, estimates of future environmental trajectories are absolutely necessary. It is of course recognized that environmental decision-making cannot be made in a vacuum. Yet scientists and public officials have a duty to inform the public about the uncertainty of environmental predictions. An honest discussion will help discourage the temptation to present dramatic scenarios and predictions of certain doom if particular environmental objectives are not met. Environmental decision-making need not be based on fear. Sound scientific and decision making discipline will result in an informed embrace that will recognize the flaws and shortcomings of the environmental predictions and in that recognition move forward to informed decisions with a new level of Christian love of the environment and reduced fear.

Editor’s note: Pictured above, President Obama meeting with Pope Francis at the Vatican March 27, 2014. (Photo credit: Pablo Martinez Monsivais / AP)

Robert V. Thomann


Robert V. Thomann is currently Professor Emeritus of Environmental Engineering at Manhattan College, Bronx, NY. He was ordained in 1977 as a permanent deacon for the Archdiocese of Newark and has ministered at Our Lady of Mount Carmel Catholic Church in Ridgewood, New Jersey. He holds a Doctor of Ministry from Fordham University (2011), a Master’s in Systematic Theology from Seton Hall University (2007), a Ph.D. in Oceanography from New York University (1963), and a Master’s in Civil Engineering from NYU (1960). He is an emeritus member of the National Academy of Engineering and has published extensively in environmental engineering and science.