Simpler models in environmental studies and predictions

ABSTRACT

This review outlines major directions of simpler model development in environmental modeling, metamodeling, statistical-regression- and machine-learning-based empirical models, and mechanistic models with reduced structures. Simpler models may be favored due to limited observational data, uncertainty in the complex model predictions, and intent of using a model as a component of a multimedia or multicompartmental model. Decision-making often relies on simple models. Model simplification can be useful in understanding the behavior of complex models. Understanding the role of models of different complexity as affected by intended uses and problem statements is an important part of the modern ontology of environmental science and technology.

KEYWORDS:

• Complexity
• environmental modeling
• metamodeling
• model simplification

Acknowledgments

Thanks are extended to Ms. Fran Rauschenberg of EPA for editing the document. We gratefully acknowledge the most helpful comments and suggestions provided by Prof. Mary Hill and five anonymous reviewers. Mention of trade names or commercial products does not constitute endorsement or recommendation for use. It has been subjected to Agency review and approved for publication.

Additional information

Funding

The United States Environmental Protection Agency (EPA) through its Office of Research and Development partially funded and collaborated in the research described here under DW-012-92348101 to the U.S. Department of Agriculture Agricultural Research Service.