Climate Change 2001: Impacts, Adaptation and Vulnerability

2.6. Characterizing Uncertainty and "Levels of Confidence" in Climate Assessment

Box 2-1. Examples of Sources of Uncertainty

Problems with Data

Missing components or errors in the data
"Noise" in data associated with biased or
incomplete observations
Random sampling error and biases
(nonrepresentativeness) in a sample

Problems with Models

Known processes but unknown functional
relationships or errors in structure of model
Known structure but unknown or erroneous
values of some important parameters
Known historical data and model structure but reasons to believe parameters or model structure will change over time
Uncertainty regarding predictability (e.g., chaotic or stochastic behavior) of system or effect
Uncertainties introduced by approximation
techniques used to solve a set of equations that characterizethe model

Other Sources of Uncertainty

Ambiguously defined concepts and terminology
Inappropriate spatial/temporal units
Inappropriateness of/lack of confidence in underlying assumptions
Uncertainty resulting from projections of human behavior (e.g., future consumption patterns or technological change), as distinct from
uncertainty resulting from "natural" sources (e.g., climate sensitivity, chaos)

Uncertainty—or, more generally, debate about the level of certainty required to reach a "definitive" conclusion—is a perennial issue in science. Difficulties in explaining uncertainty have become increasingly salient as society seeks policy advice to deal with global environmental change. How can science be useful when evidence is incomplete or ambiguous, the subjective judgments of experts in the scientific and popular literature differ, and policymakers seek guidance and justification for courses of action that could cause—or prevent—significant environmental and societal changes? How can scientists improve their characterization of uncertainties so that areas of slight disagreement do not become equated with paradigmatic disputes, and how can individual subjective judgments be aggregated into group positions? In short, how can the full spectrum of the scientific content of public policy debates be fairly and openly assessed?

The term "uncertainty" implies anything from confidence just short of certainty to informed guesses or speculation. Lack of information obviously results in uncertainty; often, however, disagreement about what is known or even knowable is a source of uncertainty. Some categories of uncertainty are amenable to quantification, whereas other kinds cannot be expressed sensibly in terms of probabilities (see Schneider et al., 1998, for a survey of literature on characterizations of uncertainty). Uncertainties arise from factors such as lack of knowledge of basic scientific relationships, linguistic imprecision, statistical variation, measurement error, variability, approximation, and subjective judgment (see Box 2-1). These problems are compounded by the global scale of climate change, but local scales of impacts, long time lags between forcing and response, low-frequency variability with characteristic times that are greater than the length of most instrumental records, and the impossibility of before-the-fact experimental controls also come into play. Moreover, it is important to recognize that even good data and thoughtful analysis may be insufficient to dispel some aspects of uncertainty associated with the different standards of evidence (Morgan, 1998; Casman et al., 1999).

This section considers methods to address such questions: first by briefly examining treatments of uncertainties in past IPCC assessments, next by reviewing recommendations from a guidance paper on uncertainties (Moss and Schneider, 2000) prepared for the TAR, and third by briefly assessing the state of the science concerning the debate over the quality of human judgments (subjective confidence) when empirical evidence is insufficient to form clear "objective" statements of the likelihood that certain events will occur.

Other reports in this collection