CADDIS: The causal analysis/diagnosis decision information system

Abstract

Biological monitoring and assessment methods have become indispensable tools for evaluating the condition of aquatic and terrestrial ecosystems. When an undesirable biological condition is observed (e.g., a depauperate fish assemblage), its cause (e.g., toxic substances, excess fine sediments, or nutrients) must be determined in order to design appropriate remedial management actions. Causal analysis challenges environmental scientists to bring together, analyze, and synthesize a broad variety of information from monitoring studies, models, and experiments to determine the probable cause of ecological effects. Decision-support systems can play an important role in improving the efficiency, quality and transparency of causal analyses. CADDIS (http://www.epa.gov/caddis) is an on-line decision framework for identifying the stressors responsible for undesirable biological conditions in aquatic systems. CADDIS was developed in response to requirements under the U.S. Clean Water Act to develop plans for restoring impaired aquatic systems. CADDIS is based on U.S. EPA's 2000 Stressor Identification Guidance document, and draws from multiple types of eco-epidemiological evidence. A major update in 2007 added summaries of commonly encountered causes of biological impairment: metals, sediments, nutrients, flow alteration, temperature, ionic strength, low dissolved oxygen, and toxic chemicals. These reviews are designed to help practitioners choose which causes to consider, based on sources, site information, and observed biological effects. A series of conceptual models illustrates connections between sources, stressors and effects. Another major new section provides advice and tools for analyzing data and interpreting results as causal evidence; these tools help quantify associations between any cause and any biological impairment using innovative methods such as species-sensitivity distributions, biological inferences, conditional probability analysis, and quantile regression analysis. An essential part of the development strategy for CADDIS has been the use of case studies to test the process and tools in different regions, and with different causal factors. Case studies have been conducted in streams on the urbanized east coast and the agriculturally-dominated mid-west to the arid west, and have considered causes including low dissolved oxygen, increased temperature, toxic substances, altered food resources and fine sediments. Lessons learned from the case studies include the importance of a structure for organizing the large variety of evidence that is often available, the need for well-matched reference sites for comparison, the benefits of iterative and directed data collection, and the frequency of surprising results. The case studies illustrate the promise of CADDIS: by building on the foundation of biological monitoring, we can provide a powerful means for improving the health of our aquatic systems. © 2009 Springer Science+Business Media, LLC.