The State of the Nation's Ecosystems: Measuring the Lands, Waters, and Living Resources of the United States. H. John Heinz III Center for Science, Economics, and the Environment. Cambridge University Press, Cambridge, United Kingdom. 2002. 288 pp., illus. $25.00 (ISBN 0521525721 paper).

Complex systems are characterized by multiple attributes with intricate interactions and emergent properties. For pragmatic reasons, it is often impossible and unnecessary to study such systems in all their dimensions. Rather, a subset of attributes is chosen for measurement, with the belief that these few variables sufficiently indicate the state of the system and allow inference to the whole. Examples of this approach include measurement of a small number of indicators to infer the state or behavior of economic systems or businesses and the health of individual human beings. However, even though these complex systems have obvious relevance to humanity, “no periodic comprehensive and reliable compilation of essential information about the overall state of the nation's environment” has been available. This void has now been partially filled, thanks in large part to a recent report on the state of major ecosystem types within the United States, commissioned by the Heinz Center for Science, Economics, and the Environment.

This ambitious undertaking identifies four broad categories of ecosystem characteristics (with accompanying indicator groupings): (1) system dimensions (extent and landscape pattern); (2) chemical and physical condition (key plant nutrients and chemical elements, selected contaminants, and physical features); (3) biological components (plant and animal species, biological communities, and ecological productivity); and (4) human use (food, fiber, water, and other goods and services provided by ecosystems). These broad categories of state variables, with accompanying indicators, are used to characterize the state of the nation as a whole, and of six major ecosystem types defined on the basis of major land-cover types (after Vogelmann et al. 2001). Indicator values specific to each ecosystem type are reported when data are available; otherwise, data needed to reliably characterize the state of a given ecosystem are scored as “indicator development needed.” In total, of the 103 indicators reported on or proposed, only 58 currently have data available, and only 31 of these proved estimates of temporal trend.

The report has an extended section of technical notes, which describe the data sources, methods of data compilation, and data limitation. One of the many strengths of the report is the attention it draws to data gaps and to indicators that are needed but that lack data. For example, farmlands and grassland/shrubland ecosystems make up about 60 percent of the land area of the United States, yet full data on these ecosystems are available for less than half of the critical indicators. Urban and suburban areas are growing most rapidly of all ecosystem types and are home to about three-quarters of all Americans, but information on these areas exists for only 6 of 15 indicators.

It is not possible here to review all the insights provided by this report, but several are worth emphasizing because of their significance to human welfare and quality of life. At the scale of the entire nation, most pronounced are changes in the extent of various ecosystem types—consistent trends are shown for freshwater wetlands (downward) and urban and suburban areas (upward), implying some causal relationship. On the basis of available data, water resources and freshwater ecosystems appear to be the most threatened and degraded. Almost every stream sampled had at least one contaminant and over 25 percent of groundwater samples had at least one contaminant that exceeded current human health standards. Alterations of the nation's water resources are most pronounced in agricultural areas—75 percent of streams sampled in farm areas exceeded federal standards for phosphorus, and 83 percent contained at least one pesticide exceeding guidelines for wildlife protection. In parallel with more comprehensive reports on the status of biodiversity in the United States (Stein et al. 2000, Stein 2002), major threats and losses were also documented for most ecosystems. For example, currently about 20 percent of native animal species (n ≈ 1700) in forest ecosystems are at risk; a similar percentage are at risk in grassland and shrub ecosystems (n ≈ 1700 species); and about 20 percent of more than 4000 native animal species that depend on streams, lakes, wetlands, or riparian areas are considered imperiled or critically imperiled (classifications in Stein 2002). However, information on the status and trend of most species (even when restricted to animals and vascular plants) is unknown.

Given the documented dependence of human societies on the goods and services provided by ecological systems (e.g., Daily et al. 1997), perhaps the most important insights provided by this report are the gaps in what is known about the condition of the nation's lands, waters, and living resources. This information gap can only be filled by devoting significant resources to long-term monitoring and assessment programs. The absence of this information will exacerbate political controversies and perpetuate the current state of affairs. Unfortunately, those segments of society most likely to oppose funding to acquire this information, because they benefit from the status quo, are in positions of political influence. However, this issue is not directly addressed by the report.

The absence of essential information also implies that ecological systems are undervalued and poorly understood or perceived to be resilient to human use and transformation. This attitude is inconsistent with recent scientific reports that document extensive human alteration of ecological systems (e.g., Vitousek 1997, Sanderson et al. 2002, Wackernagel et al. 2002), the potential for loss of essential goods and services (e.g., Baron et al. 2002), and continuing broad-scale landscape transformations as human population and per capita consumption rates grow. Combining the number of indicators that have missing data with the reality of continuing human domination of ecosystems, the necessity of an accelerated program of environmental assessment becomes obvious.

An update to The State of the Nation's Ecosystems is expected in five years—an opportunity to see if federal and state environmental agencies respond to fill in the gaps of missing information. However, even better data, in my opinion, cannot substitute for a fundamental flaw in the reporting process. The Heinz Center has adopted a policy of not attempting to identify cause-and-effect relationships because this “can influence perceptions of the scientific credibility and political neutrality of both data and reporting efforts.” This philosophy undermines the utility of monitoring data and allows only for retrospective investigations. To respond to environmental change before irreversible losses occur, it is important not only to identify or report on indicator values that trigger a critical evaluation of environmental policy but also to report on possible factors leading to such changes. As it now stands, the report relies on some unspecified individual or organization to interpret the report and assign culpability when it appears appropriate.

Other caveats, all of which can be addressed in subsequent updates, accompany this first report on the state of the nation's ecosystems. One consequence of environmental assessments conducted at broad spatial scales is the tyranny of the mean—that is, the failure to recognize significant declines in environmental quality on a local scale when broad-scale trends seem acceptable. An example is provided by forest ecosystems where the indicators of timber growth and harvest show that growth has exceeded harvest for more than 20 years. This indicator, viewed in isolation, suggests that harvest levels can be increased without adverse ecosystem effects. However, important information is masked by the value of this indicator: The density of large trees (> 73 centimeters in diameter at breast height) on public forests nationwide has decreased by 46 percent since 1952 (Powell et al. 1993); timber harvest is disproportionately focused on the largest trees; and many contemporary biodiversity issues are a consequence of the loss of large, old trees (e.g., northern spotted owls; Noon and McKelvey 1996).

Another limitation of the current report is the failure to explore the interactions among indicators. For example, in farmland ecosystems it would of interest to know the relationship between the values of the nitrates in groundwater, the phosphorus in streams, the pesticides in streams and groundwater, and the status and trends of freshwater organisms. Exploring such associations might provide insights to possible cause-andeffect relationships without sacrificing political neutrality. Finally, for indicators with inadequate data for a national assessment, the report discusses why the indicator cannot be reported at this time. However, the report does not take the next step and identify the agencies that should logically take the lead in collecting this information. As a consequence, critical information gaps may still exist when the report is reissued in five years.

Despite the caveats listed above, this report is noteworthy in initiating an essential and long overdue dialogue about the state of our nation's ecosystems. The Heinz Center and the authors of the report are to be commended for pursuing the project, documenting our current understanding of the state of the nation's environment, and attempting to insert the topic of ecosystem integrity into contemporary political discourse. Whether this report gets the attention it deserves remains to be seen.