Project Summary

Assessing the "Grand Experiment"

With the passage of the 1990 Clean Air Act Amendments, the United States embarked on an acid-deposition control policy that has been estimated to cost at least $2 billion. Title IV of the Act created a major innovation in environmental regulation by introducing market-based incentives — specifically, trading among electric utility companies in allowances to emit sulfur dioxide. The National Acid Precipitation Assessment Program (NAPAP) has been tasked by Congress to assess what Senator Moynihan has termed this "grand experiment". Developing a comprehensive assessment of the economic and environmental effects of this legislation is a major challenge. To help NAPAP face this challenge, the US Department of Energy has sponsored the development of an integrated assessment model, known as the Tracking and Analysis Framework (TAF). This paper provides a brief overview of the objectives, design, and current status of TAF.

The Objectives of TAF: A Bridge between Science and Policy

The primary objective of TAF is to help NAPAP fulfill its mandate under the 1990 Clean Air Act Amendments :

to evaluate the status of implementation, the effectiveness, and the costs and benefits of the acid-deposition control program created by Title IV of the Act, and
to determine whether additional reductions in deposition are necessary to prevent adverse ecological effects.

TAF integrates credible models of science and technology into an assessment framework that can directly address key policy issues and, in doing so, acts as a bridge between science and policy. This bridge between science and policy is designed to facilitate discussion in both directions: Policy should be informed by the best available science; and scientific research should be focused on those issues most relevant to the policy questions of primary concern. Accordingly, TAF has a set of secondary objectives:

to support coordination among scientists, to help them share, review, and assess models and data; to support communication with policy makers, about key results and insights, and

to ensure that the model reflects their concerns; and
to provide guidance for prioritizing research needs based on policy concerns and the most critical sources of uncertainty and gaps in data.

TAF is designed to be easily accessible, comprehensible, and usable so that scientists and analysts from public and private sectors can use TAF to review, understand, and contribute to the assessment. It is also flexible and extensible so that it can easily be modified to incorporate changing assumptions and to address newly emerging policy issues.

The Components of TAF

TAF links together into an integrated framework the key acid deposition components of pollutant emissions; control costs; atmospheric transport and deposition; environmental effects on visibility, lakes, soils, and human health; and valuation of these effects. TAF has been developed by a collaboration of ten different organizations, including national laboratories, universities, nonprofit organizations, and consulting firms. Each component has been developed by a different group of scientists with special expertise in those issues. Table 1 lists these components, plus the TAF management, framework design, coordination, and integration tasks, with their corresponding authors and organizations. The influence diagram on the title page also depicts the key modules with arrows representing relationships between them.

Table 1: Authors and organizations responsible for TAF Modules, integration and project management

Modules	Authors	Organizations
Emissions scenario selector	Rich Sonnenblick, Kevin Soo Hoo and Max Henrion	Lumina Decision Systems, Inc., (Lumina) Los Gatos, CA
Emissions projections and cost 1	John Molburg and Jeff Camp	Argonne National Laboratory (ANL) Argonne, IL
Emissions projections and cost 2	Jayant Kalagnanam & Stuart Siegel	Carnegie Mellon University (CMU) Pittsburgh, PA
Pathways and Deposition	Jack Shannon and Ron Marnicio	ANL, Foster-Wheeler
Visibility effects	Jack Shannon and Jeff Camp	ANL
Aquatics effects	Mitchell Small and Rajarishi Sinha Tim Sullivan	CMU E&S Environmental Chemistry, Corvallis, OR
Soils effects	Pat Ryan and Robb Turner	Oak Ridge National Laboratory
Health effects	Alan Krupnick and Dierdre Farrell	Resources for the Future (RFF) Washington, DC
Benefits	David Austin, Dallas Burtraw, Alan Krupnick, and Erin Mansur	RFF
Project management	Cary Bloyd, John Formento, and Guenter Conzelmann	ANL
System architecture, coordination and integration	Max Henrion, Rich Sonnenblick and Kevin Soo Hoo	Lumina

TAF is not a single model, but rather a flexible framework for modeling an integrated assessment. Effective integrated assessment requires progressive refinement, adaptation, and restructuring of technical models for purposes of policy evaluation. No single model will be adequate for all purposes. As science progresses, new understanding will justify revised or new models. It is therefore desirable to compare the implications of several different model formulations, based on different scientific views. As new policy questions emerge, information needs will evolve. To meet these challenges, the TAF framework is designed to accept replacements so that other modules can be slotted in to replacefor existing modules or to expand the model to address new issues. Additional modules are currently under development or planned. TAF provides a framework that allows a variety of models to be developed and coexist in a flexible, yet coordinated, manner.

TAF’s Innovative Methodology for Integrated Assessment

The successful creation of TAF in such a short time is a result of our development of an innovative methodology for integrated assessment modeling comprising a suite of interrelated methods. We believe that this methodology would be of value for other integrated assessment projects.

Too often, computer models are overly large and complicated and tend to obscure rather than illuminate the important issues. But, complexity seems unavoidable if the models are to provide credible representations of complex environmental and economic systems. TAF solves this dilemma by using reduced-form models — that is, models that are relatively simple in themselves, but derived from large, scientifically validated models. The small size of these models makes them fast to execute on a personal computer, comprehensible, and easy to modify. They derive their scientific credibility from the fact that they are constructed from and provide a demonstrably good approximation of the original detailed models.

The accessibility, transparency, and ease-of-use of TAF derives from its software architecture and user interface. The model is designed as a hierarchy of modules, created and displayed graphically as a hierarchy of influence diagrams (demonstrated on the front cover of this document). Each influence diagram shows variables and submodels as a nodes, with arrows depicting the qualitative dependencies among the variables. Behind each module and variable is a card documenting its title, units, description, mathematical definition, and other information. Highly visual organization and integrated documentation are what make TAF uniquely accessible.

TAF is implemented in Analytica, a general software modeling environment developed by Lumina Decision Systems specifically to support this kind of integrated assessment modelling. Analytica provides general facilities for hierarchical influence diagrams, integrated documentation, probabilistic analysis, and modular development.

TAF is unusual among integrated-assessment models in being developed as a collaboration between many scientists and organizations distributed over many sites across the US, as shown in Table 1. In the past, the difficulty of assembling computer models from separate components developed by different teams has made this kind of collaboration challenging at best. In this project, we adapted and refined a variety of methods from software engineering to ensure that the modules would be successfully integrated. We used a common language, provided by the Analytica modeling environment. , whose graphical influence diagrams support rapid model construction. When the project began, we first agreed on the interfaces between the modules and created a shared Public Index Library that specifies the key dimensions of arrays, such as time steps, source and receptor sites and regions, chemical species, and so on. We developed the framework and modules throughprogressive refinement with multiple iterations to identify incompatibilities, missing elements, and other integration problems. We also made extensive use of the Internet and particularly the World Wide Web, for sharing information, management, and coordination, as well as for exchanging successive module versions.

Download the complete TAF User's Guide (300+pp.) in Acrobat format (requires Adobe Acrobat)