Is Environmental Data the Missing Link?
Recent report raises questions as to how much we actually know.
By Roland Wall
Science Writer, Environmental AssociatesI. Introduction
II. What do we mean by data?
III. Using data to evaluate the
"State of the Ecosystem"
IV. How data shapes policy
V. The search for data quality
VI. Conclusion
VII. ReferencesI. Introduction
Return to top
It seems like one of the fastest ways to bore most audiences is to indicate that you're going to concentrate on "data." Immediately eyes glaze over and listeners mentally flee, anticipating dry facts, incomprehensible charts, and columns of figures that have no meaning to the unschooled eye.It is often forgotten that reliably collected data--facts, measurements and statistics, usually in numerical form--are the foundation of all modern science and technology. In the absence of data (whether about natural or human phenomena) it is impossible to make rational decisions on almost any issue. Nowhere is this more evident than when trying to evaluate the condition of the environment.
A study(1) last year by the Heinz Center for Science, Economics and the Environment underlines the importance of having accurate environmental data. In "The State of the Nation's Ecosystems," the Center's researchers identified 103 "key indicators"--measurements chosen to give a comprehensive overview of America's ecological conditions. The authors of the study were seeking to develop standardized ecological measures that can be used for monitoring the environment in much the same way that familiar economic indicators (such as the unemployment rate) are used to evaluate the economy.
Yet, in attempting to assess these indicators, the study reports an unexpected finding: for almost half of the chosen measurements, accurate data does not yet exist to evaluate the issue in question. Put another way, the best science currently available can only accurately judge about half of the nation's important ecological variables. This proved to be true for environmental factors as disparate as groundwater levels, coastal erosion and agricultural soil salinity. Over 40 such areas in all could not be reliably evaluated.
In this issue, we will use the results of the Heinz Report as a focus for demonstrating the importance of data in understanding and protecting the natural and human environments. We will first consider what is meant by data, how it is collected, and some of the impacts data has on environmental policy. Finally, we will consider some of the conflicts that arise in determining how data gets used.
It is far too easy to see a discussion of data as something abstract or theoretical. Data, however, are nothing more nor less than descriptions of reality. Translating real phenomena into a measurable form allows us to better understand and influence outcomes. In order to preserve, to protect or to change environmental conditions, we must first have an objective understanding of those conditions. Data allows us that understanding.
II. What do we mean by data?
Return to top
First, it should be emphasized that "data" is not an exotic or esoteric topic. The process of gathering data and applying it to decision- making is a normal part of our daily lives.
Let's take a simple example. One of the most common pieces of environmental data used in society is the "ambient air temperature," i.e. whether its hot or cold outside. Before we step outside in the morning, most of us make some attempt to assess this particular data, even it we don't realize we're doing it. We may, for example, simply stick our head out the door and see how the apparent temperature fits our comfort level. This is a crude but effective method of data collection.
On the other hand, a large number of people look at a thermometer or listen to a weather report, gaining a numerical piece of data--the morning temperature. That numerical data then enters our decision making process--a temperature of thirty degrees Fahrenheit, for example, will lead us to different choices (clothes, travel plans, etc) than a temperature of seventy.
By taking raw data and applying it to decisions, we are, in fact, engaging in "information processing." While data is the raw material, information is derived from placing data in context, aggregating it or otherwise analyzing it. The morning temperature has meaning simply because we can compare it to different temperatures. On the other hand, when we look at more complex environmental issues, the types of information needed to make decisions may require more sophisticated analysis. Nonetheless, it is the quantity and quality of the data which makes the analysis and the decisions possible.
This is really all that "data" means: facts, collected objectively, to provide information that is then used to guide rational decisions. And just as decisions can vary from the simple to complex, so too can data run the gamut from an indicator as simple as the air temperature to measurements as complex as long term climate trends.
For those who live and work in immediate contact with natural systems - e.g. farmers, ranchers, foresters--there is a broad range of environmental indicators which must be assessed on a regular basis. Plant growth rates, soil nutrient levels, and groundwater availability are the types of measures which can make a critical difference for people in these sorts of professions.
For those inhabiting industrialized settings, the use of environmental data may seem more distant and vague. Yet, cities are ecosystems too, and even the most jaded urbanite walking up the street will be noting landscape features, judging the topography, listening to bird songs or swatting mosquitoes. Anywhere we are interacting with the environment, we are--in a sense--collecting personal environmental data to guide our actions.
But if environmental data collection on an individual scale is commonplace, aggregating such data on a national and global level is still in its infancy. This contrasts starkly with economic information, for which data is constantly being collected, and for which there are well developed systems of indexing and reporting. From local interest rates to measures of gross domestic product, economic data is widely available on both micro and macro scales. Elections, stock prices, and livelihoods can all be decided by the outcome of these analyses. Ironically, for measuring ecological conditions (which can also be literally a matter of life and death), society as yet does not have analogous standards and measures.
What sort of data might one collect that would tell us as much about the ecosystem as economic indicators tell us about the health and future of the economy? Answering this question is a key step in protecting the environment, and was a major portion of the Heinz Center study. Starting with a mandate from the Federal government in 1997, the Center assembled over 150 of the nation's top environmental professionals, including diverse representatives of industry, government, advocacy organizations, and academia. (Many other experts would ultimately join in the process.)
These specialists first worked to define the indicators that would best reflect the current state of the ecosystem. This was an arduous task, and required the expertise and judgment of the full range of interests and disciplines that were represented, as well as significant discussion and compromise. Dr. Thomas Lovejoy, President of the Heinz Center, noted at the time the report was released, "our participants-including many traditional adversaries-put aside their differences to agree on scientifically grounded and policy-relevant indicators for describing the state of our natural systems." (2)
Because the nation is a patchwork of different land uses and geographic features, researchers divided the country into six basic, broadly-drawn ecosystem types: forest, farmland, coast and oceans, freshwater, scrubland, and urban/ suburban development (see Box 1). These six classes provide a wide scale of reference for visualizing the nation's ecological processes. (Some measures were also chosen that would indicate ecological conditions for the nation as a whole.)
Box 1: Defining Selected Ecosystems
Coasts and Oceans: Estuaries and ocean waters under U.S. jurisdiction. Estuaries are partially enclosed bodies of water generally considered to begin at the upper end of tidal or saltwater influence and end where they meet the ocean. By definition, U.S. waters extend to the boundaries of the U.S. Exclusive Economic Zone (EEZ), which extends 200 miles from the U.S. coast.
Farmlands: This indicator covers both croplands-"lands used for production of annual and perennial crops and livestock"-and the overall landscape of farm regions--"field borders and windbreaks, small woodlots, grassland or shrubland areas, wetlands, farmsteads, small villages and other built-up areas within and adjacent to croplands."
Forests: The US Forest Service defines of forest as "lands at least 10% covered by trees of any size, at least one acre in extent. This includes "both naturally regenerating forests and areas planted for future harvest (plantations or 'tree farms'). "
Fresh Waters: This includes rivers and streams;, lakes, ponds, and reservoirs; groundwater; freshwater wetlands; riparian areas-vegetated margins of streams and rivers (and including the edges of lakes.) Freshwater is further defined by the Water Environment Federation as water "containing an insignificant amount of salts."
Grasslands and Shrublands: "Lands in which the dominant vegetation is grasses and other nonwoody vegetation, or where shrubs (with or without scattered trees) are the norm."
Urban and Suburban Areas: Those places where the land is primarily devoted to buildings, houses, roads, concrete, grassy lawns, and other elements of human use and construction.
Adapted from Heinz Center, 2002(a).
Ten areas were assessed for each ecosystem (see Box 2). These included geographic characteristics (e.g. size, shape, pattern, fragmentation), chemical composition (nutrients, contaminants), physical conditions (erosion, groundwater depth), biological factors (biodiversity, productivity), and "ecosystem services" (i.e. valuable services such as recreation or water purification that depend on the ecosystem's functions).
Box 2 - Groups of Indicators
Ecosystem extent – Gains or losses in the area covered by a particular ecosystem.
Fragmentation and landscape pattern – Size, shape, proximity and other patterns of how ecosystems are arranged on the landscape.
Building blocks of life – Amounts and concentrations of key chemicals (nitrogen, phosphorous, carbon, and oxygen) that play vital roles in ecosystems.
Contaminants – The extent of chemical contamination, as well as the frequency with which contaminant levels exceed regulatory standards and advisory guidelines.
Physical conditions – The condition of important physical characteristics of a particular ecosystem, such as coastal erosion or the depth to groundwater.
Plants and animals – The presence and condition of native and nonnative species of plants and animals.
Biological communities – The condition of groups of plants and animals that form the "biological neighborhood" for other species.
Plant growth and productivity – The amount of plant growth, which reflects the amount of energy entering an ecosystem and available to all organisms.
Production of food and fiber and use of water – Quantities of goods produced by ecosystems, such as crops, livestock, timber, fish, and water.
Recreation and other services – Activities like swimming, hiking, biking, and hunting, and other services, including plant pollination and flood reduction.
Adapted from Heinz Center, 2002(a).
Under the ten broad categories, a total of 103 measurements and indicators were chosen to be collected. These are specific to each ecosystem type, and range from such obvious concerns as "Pesticides in Farmland Streams and Groundwater" to more subtle measures--e.g. "Population Trends in Invasive Birds." Like economic measures, each tells something specific about the state of the system, and like economic measures, none tell the whole story.
All this was done prior to determining exactly which data could be accessed. As one leader of the effort noted "we sought to define what should be available, then determine what in fact was available." The next step in the process, therefore, was to examine a variety of datasets to determine which information already existed that would describe the indicators. The project did not generate new data--i.e. did not make any new measurements--but rather sought to find the needed material in the mass of data that has already been collected for various applications.
Three criteria were used to judge which sets of data should be utilized. First, they had to be scientifically credible, meeting "the highest standards of the appropriate discipline." Second, they had to be applicable to a large portion of the ecosystem being examined. If, for example, the data applied to forests, it should be something measured over a large proportion of the forests in the nation. Finally, because the purpose of the study was to provide a baseline of information to be tracked for years or decades, data had to have been collected as part of an on-going program.
(These three factors provide good criteria in general for judging environmental data. Are they credible? Are they representative of the total system? Do they provide a means of judging trends over time? In this sense the Heinz report is a model for evaluating how data could be chosen for decision making.)
Of the measures chosen, 43 have not had adequate data collected to be assessed. In most cases, this was simply because no one had yet bothered to do so, or at least have not assembled the data in a usable form. In a smaller number of cases, scientists have not yet determined the best way to measure the indicator. The study called for "a multisector effort is needed to address key gaps identified in this report."
III. Using data to evaluate "The State of the Ecosystem"
Return to top
Despite these gaps, the Heinz Center report did assemble important information about the current state of ecosystems in the United States. While the results are too extensive to fully summarize here, some examples serve to demonstrate how data can be utilized to better understand environmental conditions:
- Movement of Nitrogen: The element nitrogen is the major component of the Earth's atmosphere; as a part of chemical compounds it is the foundation for many vital biochemical and ecological processes. Its distribution is the result of a complex natural cycle which at any given time divides it among the planet's atmosphere, water, soil and living things.
Nitrate, a common form of nitrogen is present in freshwater and is used as a nutrient by aquatic plants. It is also a vital ingredient of agricultural fertilizers. In excess amounts, however, nitrates can lead to harmful algae "blooms" and to oxygen depletion in freshwater systems. Because nitrogen movement is indicative of human-influenced ecological change, it is used as one of the core national indicators of ecosystem condition.
To assess the movement of nitrogen through the ecosystem the Heinz study looked at both nitrate and (where available) total nitrogen in the rivers of the nation's major watersheds. These measurements reflect the degree to which the land is "leaking" nitrogen into the waterways. The amounts of nitrogen were found to have steadily increased in recent decades. The Mississippi River for example transported less than 500 million pounds of nitrogen in 1950; by the year 2000 that amount had risen to over 1 billion.
- Sea Surface Temperature: The temperature of the sea at it's surface is thought to be related to a variety of factors affecting marine systems. In particular the health of coral reefs and the development of harmful algae are impacted by increases in surface temperature. Overall, the biological makeup of shore and coastlines directly reflects the fluctuations in surface temperatures.
Interestingly, this particular indicator showed no significant changes over time, though it varied widely from year to year. This is an example of data being important even when it does not reveal any dramatic human impact on the environment. By providing a baseline--an indication of "normal" conditions--the data can be used as a frame of reference for analyzing future trends. A wide body of reference data is just as crucial as data that shows changes.
- Pesticides in Farmland Streams and Groundwater: It is self-evident that the presence of pesticides in surface and groundwater would be source of concern from a public health standard. As an ecological measure, the presence of excess pesticides in the environment suggests the impact farming is having on natural systems. However, the report emphasizes, presence of the pesticide alone does not necessarily indicate problems. It is when pesticides are present in amounts that exceed recommended environmental and health standards that they become an environmental issue.
In this study it was found that all of the streams monitored in farm areas had "at least one pesticide at detectable levels throughout the year, and about 75% had an average of five or more." In 83% of the streams, at least one pesticide was present in amounts exceeding aquatic life guidelines. In 4% of the cases, human health standards were also exceeded. Groundwater fared better. Although 60% of farm wells had pesticide at detectable levels, less than 1% exceeded health standards.
- Patches of Forest, Grasslands and Shrublands, and Wetlands: This indicator relates to the amount of natural areas left within urban and suburban ecosystems. The sizes and shapes of undeveloped patches of land are critical factors in determining their value as habitat. This particular measure looked only at the size of the patches, data that are easily obtained from satellite surveys.
Over half of all natural areas in urban regions are of less than ten acres in size. The number of larger patches decreases progressively as the size of the patch increases. In other words, only a tiny fraction of urban wooded areas are of a size greater than a hundred acres, and even fewer of greater than a thousand. The implication of this is that "natural" urban areas are most likely to support tolerant, "edge" species that are well adapted to human influence. (This, however, is a generalization and may be influenced by shape and location of the patch--information that is not as readily available.)
* * *
These are just a few examples of the indicators the Heinz report applies to assess ecological conditions. In considering the use of data for decision-making, however, we must also take into account issues related to the quality of data and how it is obtained.
For every piece of data considered there are many others that are not taken into account. This is because it would be impossible to consider every single data point. Nor is every piece of data equally useful. According to the Heinz Center, the goal of the report is to identify a "succinct set" of "strategic indicators that can serve as meaningful reference points for broad-ranging policy discussions."
At the same time it should be recognized that obtaining such data is not a trivial process. There is no central body that gathers and analyzes environmental data. To some extent the Heinz report represents a cobbling together of information from a multitude of sources.
Thus the data on nitrogen movement was derived from several nationwide stream quality surveys conducted by the US Geologic Survey, while the sea temperature was the result of data collected from orbiting satellites operated the National Oceanic and Atmospheric Administration. And the urban land use data, also a product of satellite remote sensing, was the result of group of organizations known as the Multi-Resolution Land Characterization Consortium (MRLC).
The point here is that the sort of broad, varied data required to assess the nation's ecosystem can only be obtained from multiple sources. That in turn requires expert determination of data quality -- its validity, relevance and accuracy. These questions, can become mired in both technical and ideological conflicts.
While the Heinz study put great emphasis on valid, unbiased choices of data categories and specific data sets, they acknowledge that "the values held by different people can lead them to place greater importance on some aspects of ecosystems than on others." Although for this particular study there was a generally agreed upon procedure of negotiation and peer review, such a framework does not always exist when applying data. The fact that selection of indicators is, in the words of the study, "inevitably a value-driven process" can lead to significant controversy.
Finally, as has already been mentioned, the study was limited by "critical gaps" that made it impossible to obtain data on many of the indicators. These gaps lessen the conclusions that can be drawn from the data that is available.
For example, both the measures of nitrogen movement and of the sea surface temperature are related to the harmful algae "blooms" that can devastate aquatic systems. Yet no data are available for measuring the algal blooms themselves, another key indicator that was identified by the report. Indeed, there does not even exist a generally agreed upon method for making such measurements or for determining if such blooms are more or less frequent than in the past.
Filling data gaps will be one of the most important roles of ecological research in coming decades. As the report writers note, they sought to identify "where lamps need to be posted in order to provide the kind of illumination of ecosystems that the nation most needs." For the present, it is sobering to learn that our best environmental intentions may be meaningless if we cannot develop a better understanding of environmental realities.
IV. How data shapes policy
Return to top
Simply having data, however, does not mean that environmental problems will be solved. The data must be applied to individual and collective decision-making. In the US this is often a political process, involving the interaction of government, economic interests and public opinion. To better understand how data informs public debate and guides society's actions towards the environment, it is helpful to distinguish between three different terms that are often confused in discussions of environmental issues--ecology, environmentalism and environmental policy.
Ecology is the branch of science that studies the relationship between organisms and their environment. Like all sciences, it is based on provable hypotheses that are tested by experiments that yield credible results. And like all sciences, conclusions must be supported by reproducible evidence--that is, by data. Ecology is not a political ideology, nor does it necessarily relate to human impacts on the environment.
Environmentalism, loosely defined, is a political ideology, or more accurately, a political viewpoint. One dictionary defines environmentalism as "advocacy for, or work toward, protecting the natural environment from destruction or pollution." Environmentalists often make use of the science of ecology and the data that is associated with it, but they may not be as rigorous in applying data as scientists might consider necessary.
The broader term environmental policy, in general, refers to laws, economic practices, and governmental actions and procedures that impact on the environment. In democratic societies, environmental policy is the result of social and political actions (ranging from voting and lobbying to litigation and demonstration) by various stakeholders. (Though, as is obvious to anyone who has observed policy making, depending on the issues and circumstances, some stakeholders have more power and influence than others.)
What does all this have to do with data? A great deal, actually. Environmental policy, more than many areas of government or administration, is directly related to concepts that have objective scientific validity. Ecology (along with other branches of science) produces and analyzes the data that allows democratic societies to make decisions on environmental matters. Although values may differ from stakeholder to stakeholder (including environmentalists), the assumption is that the better the quality of the data used, and the more accurate our interpretations of it, the more likely it is that environmental policy will be made in the best interests of both human and natural systems.
Many of the environmental policies we've discussed in the past (the Clean Water Act, the Safe Drinking Water Act, the Toxic Substances Control Act, among others) are the direct result of--and remain dependent on--data collection. From the pioneering ecosystem studies by the Academy of Natural Sciences that facilitated the passage of the Clean Water Act, to the present, ongoing research and monitoring that guides the supplying of safe drinking water, environmental policy is often driven by the results of data collection.
For example, in our discussion of polychlorinated biphenols (PCBs--see KYE 2/2002), we saw that data collected in the late 1960's revealed--with a high level of reliability--that traces of these once-widely used industrial chemicals could be found in animal tissue and ecosystems around the world. The discovery came as a shock to both politicians and the public; the passage of the federal Toxic Substances Control Act (TSCA) was a direct result of that data.
Unfortunately, though, despite the best efforts of science to produce data and analysis which are objective, rational, and unbiased, environmental policy-making rarely has the luxury of complete certainty. In part, this is because of the gaps in data such as those the Heinz report revealed, including gaps in our ability to even derive ways of measuring certain phenomena. But the problem also lies in the level of complexity of most environmental issues, in the need to meld the inherent uncertainties of natural science with the cascade of competing--legitimate--interests that exist in society.
Thus, thirty years after the passage of the TSCA, debate continues as to the actual effects of PCBs in the environment, with data analysis and conclusions often influenced by the particular interests presenting them. Differences in scientific disciplines, ideologies, perceptions, and judgments as to credibility can all affect how data may be applied to environmental debates. In comparing reputable studies of PCBs in the environment, for example, we noted that it is possible for two sets of equally plausible data to be presented in ways that completely contradict each other.
This may, in part, be due to differences in approaches to assessing environmental consequences. "Risk based"assessments, for example, attempt to define acceptable levels of impact and to analyze, often statistically, the likelihood that a given practice or policy will result in such impacts.The "precautionary principle" on the other hand, essentially states that all practices impacting the environment--the use of artificial substances--should be assumed to have harmful consequences until proven otherwise. While both of these approaches may have validity in particular cases, they may also be seen as representative of specific interest groups, and cannot always be defended strictly on technical merit.
In some cases, there is a concerted effort to develop unbiased analysis which is beyond reproach. The National Academy of Sciences for example, working through such divisions as the National Research Council, often serves to provide studies that evaluate and balance conflicting scientific judgements. Their work is highly respected, though at times the focus may seem narrower than particular stakeholders would prefer. This is because it is rare for environmental issues to be solely a matter of science.
For that reason, ecology alone will never be the final arbiter of environmental policy. Stakeholders--some who define themselves as environmentalists, some who represent a variety of other interests--will always be prone to using data in ways that promote advocacy. This is true for all political and policy issues, but is somewhat more critical on environmental issues because the science is so central to most debates.
V. The Search for Data Quality
Return to top
All this is not to diminish the importance of data quality for environmental decision making. Rather, it is to emphasize that having accurate data is a "necessary-but-not-sufficient" condition for making sound environmental policy. This point has come to the forefront in recent months in the debate over the federal Data Quality Act (DQA)--a little-noticed piece of legislation that could potentially have great impact on how data is used to make environmental policy.
The DQA was passed during the waning days of the Clinton Administration as a rider to a budget bill. There were no hearings on it, and the wording of the Act was brief--just 27 lines. And though it has led to controversy, the intent of the legislation could not seem more benign--"ensuring and maximizing the quality, objectivity, utility and integrity" of the scientific data that government agencies disseminate and use for policy making.
The Act requires that each federal agency issue guidelines (based, in turn, on guidelines from the Office of Management and Budget) that would ensure data used by the agency meets clear quality standards. It also requires that each agency develop steps that "affected persons" can take "to seek and obtain correction" of data they (the affected persons) believe to be in error. For data that does not meet the standards, such a correction would require removal of the data from government publications and websites, and adjustment of any regulations that are based upon it.
At first glance it is difficult to understand why anyone would object to a measure that seeks to improve the quality of the data used to make regulatory decisions. Jim Tozzi, a former OMB official and current board member of the Center for Regulatory Effectiveness (an advocacy group that played a prominent role in the passage of the Act) notes in a recent editorial (3) that "increased confidence in the scientific information available to the government will eliminate one complication" in policy decisions.
Furthermore, Tozzi suggests several other benefits that will derive from new standards for data quality. The law, according to Tozzi will codify the requirement that data used and disseminated by the federal government be "objective, unbiased, transparent, and reproducible," allowing scientists to know "with some certainty the quality standards against which the information they generate will be judged."
The law will also apply to "third party petitions or other information submissions to federal agencies." Tozzi states that this "requirement will apply to all types of advocacy groups, both from industry and environmentalists," and would create "a level playing field." Finally, because the law will instill confidence in "the regulated community that agency rules and pronouncements have a rational basis in science" there will be fewer legal and administrative challenges to regulations.
Despite these arguments, and despite the fact that the Data Quality Act will apply to all branches of government, some have criticized it as being aimed at delaying or derailing environmental regulations. Sen. Jim Jeffords (I-VT), former chair of the Senate Environment and Public Works Committee, has been outspoken in his criticisms of the Act, stating that its supporters "have latched on to the Data Quality Act and are attempting to misuse it to prevent the public from getting valid information about threats to their well being and quality of life."
Chuck Herrick, a former staff member of the White House Council on Environmental Quality, wrote in an editorial response (4) to Mr. Tozzi, "It strikes me that the rationale for DQA rests upon a fundamental misunderstanding concerning the nature of scientific assessment in a policy context." According to Herrick, policies "are typically based upon a wide variety of informational inputs, some of which are more robust than others." Policy makers should "consider the entire mosaic," not "zero-in on a few individual tiles."
Herrick further contends that "it is probably unwise to pre-stipulate absolute measures of acceptable data quality. Rather, the value of a particular data set is determined by its 'fitness for use' in a particular situation." He also states a concern that the removal of the data from government publications based on preset standards would diminish the opportunity for the normal process of scientific inquiry and consideration. "[I]t is only through publication and dissemination of findings and alternative scenarios that meaningful research dialog can occur." Withdrawing material produced by reputable experts "would retard both research and policy deliberation, assuring only an ignorant status quo."
The Data Quality Act was scheduled to take effect in late 2002, though it will no doubt be subject to administrative and judicial action for some time to come. How it will ultimately affect the use of data by the government is still uncertain. (In a bizarre twist, an early application of the Act has been by a group filing for National Aeronautics and Space Administration (NASA) to "correct" data and to support a claim of intelligent life on Mars, suggesting that the Act may be applied in some highly unexpected ways.)
Although the DQA applies to all government agencies (even, as we see, NASA), it is probably accurate to say that environmental regulation motivated its passage. One of the first filings under the Act was by Mr. Tozzi's organization calling for withdrawal of a recent report on global climate change; other organizations appear prepared to file actions on environmental regulations as soon as the data guidelines are finalized for various agencies.
Despite the controversy over this latest attempt to improve the quality of data used for environmental policy, there is no doubt that everyone involved in policy making and scientific research supports the principle that accurate, credible data be applied to decisions. Mr. Herrick acknowledges that if it is applied flexibly, recognizing "the idiosyncratic nature of information use in a policy context," and is not treated as a "codification of immutable rules" - then the Act could have positive consequences. By emphasizing data correction, the DQA could help make "agencies less bureaucratic, more open and transparent, and more responsive to citizen input."
VI. Conclusion
Return to top
In light of these political twists, the Heinz Center report takes on new meaning. If the government is now holding data to very high standards of quality in order for it to be applied in policy making, it is all the more critical that environmental data be obtained in those areas that are currently lacking. This is a tremendous research and organizational project.
Moreover, it is important that mechanisms be developed for gathering and interpreting data that will raise fewer questions as to data quality. As Dr. Lovejoy concluded at the release of the Heinz report "Policymaking about the environment will always be contentious in a democracy. But debates on how best to manage our nation’s natural resources should not be sidetracked through needless debates about the facts." (5)
Thus, in economics, politicians may argue over the reasons for the current unemployment rate but they rarely argue as to whether or not the unemployment data itself is valid or biased. Systems are in place that give policy makers a high degree of confidence on economic data. Similar systems would help lessen uncertainty on environmental issues.
Nevertheless, uncertainty will remain a fact of life in environmental affairs. Given that the Heinz study suggests that data on some of their recommended indicators may not be forthcoming in the near future, policy makers will continue to grapple with decision-making in the absence of facts. It will be all the more important under those conditions to distinguish between high quality data, and data that has been selectively acquired to advance advocacy agendas. In a democratic society, both types of data are important, however, it is equally important that they not be confused.
Finally, decision makers should remember that, although data may be gathered and analyzed in a detached and rational setting, the data itself reflects processes that are occurring in the real world. Factors such as pesticide in drinking water or changes in land use patterns may be represented by numbers or charts, but in reality they represent impacts on individual organisms, on human and biological communities, and on entire ecosystems. Scientists, policy makers and the public would do well to realize that decisions based on data, however arcane, ultimately translate back to the real world as actions and outcomes.
VII. References
Return to top
1. Heinz Center, 2002 (a). The State of the Nation’s Ecosystems: Measuring the Lands, Waters, and Living Resources of the United States. The H. John Heinz III Center for Science, Economics and the Environment . Cambridge University Press, New York. Note: All quotes in the article are from this report unless cited otherwise.2. Heniz Center, 2002(b). "Heinz Center Issues Landmark Report on State of the Nation's Ecosysems: Unique Collaboration Presents Key Environmental Indicators and Identifies Gaps." The H. John Heinz III Center for Science, Economics and the Environment. Press Release: Sept. 24, 2002.
3. Tozzi, J. 2002. "The Data Quality Act: A New Tool for Ensuring Clarity at the Interface of Science and Policymaking." Guest Editorial. Ogmius, No. 2, May 2002. Center for Science and Technology Policy Research: Univ. of Col.-Boulder.
4. Herrick, C. 2002. "Ogmius Exchange: Editorial Response." Ogmius, No. 2, May 2002. Center for Science and Technology Policy Research: Univ. of Col.-Boulder.
©1998 - 2004 All Rights Reserved, The Academy of Natural Sciences, 1900 Benjamin Franklin Parkway, Philadelphia, PA 19103
Tel: 215-299-1000 Email:
