This article identifies conditions under which robust predictive modeling results have special epistemic significance---related to truth, confidence, and security---and considers whether those conditions hold in the context of present-day climate modeling. The findings are disappointing. When today’s climate models agree that an interesting hypothesis about future climate change is true, it cannot be inferred---via the arguments considered here anyway---that the hypothesis is likely to be true or that scientists’ confidence in the hypothesis should be significantly increased or that a claim (...) to have evidence for the hypothesis is now more secure. (shrink)

To study Earth’s climate, scientists now use a variety of computer simulation models. These models disagree in some of their assumptions about the climate system, yet they are used together as complementary resources for investigating future climatic change. This paper examines and defends this use of incompatible models. I argue that climate model pluralism results both from uncertainty concerning how to best represent the climate system and from difficulties faced in evaluating the relative merits of complex models. I describe how (...) incompatible climate models are used together in ‘multi-model ensembles’ and explain why this practice is reasonable, given scientists’ inability to identify a ‘best’ model for predicting future climate. Finally, I characterize climate model pluralism as involving both an ontic competitive pluralism and a pragmatic integrative pluralism. (shrink)

We argue that concerns about double-counting—using the same evidence both to calibrate or tune climate models and also to confirm or verify that the models are adequate—deserve more careful scrutiny in climate modelling circles. It is widely held that double-counting is bad and that separate data must be used for calibration and confirmation. We show that this is far from obviously true, and that climate scientists may be confusing their targets. Our analysis turns on a Bayesian/relative-likelihood approach to incremental confirmation. (...) According to this approach, double-counting is entirely proper. We go on to discuss plausible difficulties with calibrating climate models, and we distinguish more and less ambitious notions of confirmation. Strong claims of confirmation may not, in many cases, be warranted, but it would be a mistake to regard double-counting as the culprit. 1 Introduction2 Remarks about Models and Adequacy-for-Purpose3 Evidence for Calibration Can Also Yield Comparative Confirmation3.1 Double-counting I3.2 Double-counting II4 Climate Science Examples: Comparative Confirmation in Practice4.1 Confirmation due to better and worse best fits4.2 Confirmation due to more and less plausible forcings values5 Old Evidence6 Doubts about the Relevance of Past Data7 Non-comparative Confirmation and Catch-Alls8 Climate Science Example: Non-comparative Confirmation and Catch-Alls in Practice9 Concluding Remarks. (shrink)

In computer simulations of physical systems, the construction of models is guided, but not determined, by theory. At the same time simulations models are often constructed precisely because data are sparse. They are meant to replace experiments and observations as sources of data about the world; hence they cannot be evaluated simply by being compared to the world. So what can be the source of credibility for simulation models? I argue that the credibility of a simulation model comes not only (...) from the credentials supplied to it by the governing theory, but also from the antecedently established credentials of the model building techniques employed by the simulationists. In other words, there are certain sorts of model building techniques which are taken, in and of themselves, to be reliable. Some of these model building techniques, moreover, incorporate what are sometimes called “falsifications.” These are contrary-to-fact principles that are included in a simulation model and whose inclusion is taken to increase the reliability of the results. The example of a falsification that I consider, called artificial viscosity, is in widespread use in computational fluid dynamics. Artificial viscosity, I argue, is a principle that is successfully and reliably used across a wide domain of fluid dynamical applications, but it does not offer even an approximately “realistic” or true account of fluids. Artificial viscosity, therefore, is a counter-example to the principle that success implies truth – a principle at the foundation of scientific realism. It is an example of reliability without truth. (shrink)

A lecture before the Harvard Natural History Society; published in the Atlantic Monthly; and later republished in James (1897)The Will to Believe and Other Essays in Popular Philosophy.

The sensitive dependence on initial conditions (SDIC) associated with nonlinear models imposes limitations on the models’ predictive power. We draw attention to an additional limitation than has been underappreciated, namely, structural model error (SME). A model has SME if the model dynamics differ from the dynamics in the target system. If a nonlinear model has only the slightest SME, then its ability to generate decision-relevant predictions is compromised. Given a perfect model, we can take the effects of SDIC into account (...) by substituting probabilistic predictions for point predictions. This route is foreclosed in the case of SME, which puts us in a worse epistemic situation than SDIC. (shrink)

In 1995, the Intergovernmental Panel on Climate Change announced that anthropogenic climate change had become discernible. Since then, numerous independent studies have affirmed that anthropogenic climate change is underway, and the meta-conclusion that there is a broad expert consensus on this point. It has also been demonstrated that most of the challenges to this claim come from interested parties outside the scientific community. But even if we allow that the challenges to climate science are politically or economically motivated, it does (...) not prove that the scientific consensus is correct. In other words, even if we accept the fact of scientific consensus, how do we know that this consensus is not wrong? This chapter addresses this question by examining a set of criteria that philosophers have traditionally or recently identified as possible bases for trust in scientific conclusions, and shows that climate science meets all of these criteria. Thus, while there is no way to know for sure that scientists are correct in their conclusions, the various means we have to test and evaluate scientific claims lead to the conclusion that, so far as we are able to tell, it is most likely that scientists are not wrong about the reality of anthropogenic climate change. (shrink)

I bring out the limitations of four important views of what the target of useful climate model assessment is. Three of these views are drawn from philosophy. They include the views of Elisabeth Lloyd and Wendy Parker, and an application of Bayesian confirmation theory. The fourth view I criticise is based on the actual practice of climate model assessment. In bringing out the limitations of these four views, I argue that an approach to climate model assessment that neither demands too (...) much of such assessment nor threatens to be unreliable will, in typical cases, have to aim at something other than the confirmation of claims about how the climate system actually is. This means, I suggest, that the Intergovernmental Panel on Climate Change ’s focus on establishing confidence in climate model explanations and predictions is misguided. So too, it means that standard epistemologies of science with pretensions to generality, e.g., Bayesian epistemologies, fail to illuminate the assessment of climate models. I go on to outline a view that neither demands too much nor threatens to be unreliable, a view according to which useful climate model assessment typically aims to show that certain climatic scenarios are real possibilities and, when the scenarios are determined to be real possibilities, partially to determine how remote they are. (shrink)

Simulation-based weather and climate prediction now involves the use of methods that reflect a deep concern with uncertainty. These methods, known as ensemble prediction methods, produce multiple simulations for predictive periods of interest, using different initial conditions, parameter values and/or model structures. This paper provides a non-technical overview of current ensemble methods and considers how the results of studies employing these methods should be interpreted, paying special attention to probabilistic interpretations. A key conclusion is that, while complicated inductive arguments might (...) be given for the trustworthiness of probabilistic weather forecasts obtained from ensemble studies, analogous arguments are out of reach in the case of long-term climate prediction. In light of this, the paper considers how predictive uncertainty should be conveyed to decision makers. (shrink)

The United Kingdom Climate Impacts Programme’s UKCP09 project makes high-resolution projections of the climate out to 2100 by post-processing the outputs of a large-scale global climate model. The aim of this paper is to describe and analyse the methodology used and then urge some caution. Given the acknowledged systematic, shared errors of all current climate models, treating model outputs as decision-relevant projections can be significantly misleading. In extrapolatory situations, such as projections of future climate change, there is little reason to (...) expect that post-processing of model outputs can correct for the consequences of such errors. This casts doubt on our ability, today, to make trustworthy probabilistic projections at high resolution out to the end of the century. (shrink)

The United Kingdom Climate Impacts Program’s UKCP09 project makes high-resolution forecasts of climate during the 21st century using state of the art global climate models. The aim of this paper is to introduce and analyze the methodology used and then urge some caution. Given the acknowledged systematic errors in all current climate models, treating model outputs as decision relevant probabilistic forecasts can be seriously misleading. This casts doubt on our ability, today, to make trustworthy, high-resolution predictions out to the end (...) of this century. (shrink)

This article examines initial-condition dependence and initial-condition uncertainty for climate projections and predictions. The first contribution is to provide a clear conceptual characterization of predictions and projections. Concerning initial-condition dependence, projections are often described as experiments that do not depend on initial conditions. Although prominent, this claim has not been scrutinized much and can be interpreted differently. If interpreted as the claim that projections are not based on estimates of the actual initial conditions of the world or that what makes (...) projections true are conditions in the world, this claim is true. However, it can also be interpreted as the claim that simulations used to obtain projections are independent of initial-condition ensembles. This article argues that evidence does not support this claim. Concerning initial-condition uncertainty, three kinds of initial-condition uncertainty are identified. The first is the uncertainty associated with the spread of the ensemble simulations. The second arises because the theoretical initial ensemble cannot be used in calculations and has to be approximated by finitely many initial states. The third uncertainty arises because it is unclear how long the model should be run to obtain potential initial conditions at pre-industrial times. Overall, the discussion shows that initial-condition dependence and uncertainty in climate science are more complex and important issues than usually acknowledged. 1Introduction2Projections and Predictions 2.1Predictions2.2Projections3Initial-Condition Dependence 3.1Projections 3.1.1Dynamical conditions justifying independence of initial conditions?3.1.2What projections can and cannot provide3.2Predictions4Initial-Condition Uncertainty 4.1Projections4.2Predictions5ConclusionAppendix. (shrink)

An uncertainty report describes the extent of an agent’s uncertainty about some matter. We identify two basic requirements for uncertainty reports, which we call faithfulness and completeness. We then discuss two pitfalls of uncertainty assessment that often result in reports that fail to meet these requirements. The first involves adopting a one-size-fits-all approach to the representation of uncertainty, while the second involves failing to take account of the risk of surprises. In connection with the latter, we respond to the objection (...) that it is impossible to account for the risk of genuine surprises. After outlining some steps that both scientists and the bodies who commission uncertainty assessments can take to help avoid these pitfalls, we explain why striving for faithfulness and completeness is important. (shrink)

This article argues that common intuitions regarding (a) the specialness of ‘use-novel’ data for confirmation and (b) that this specialness implies the ‘no-double-counting rule’, which says that data used in ‘constructing’ (calibrating) a model cannot also play a role in confirming the model’s predictions, are too crude. The intuitions in question are pertinent in all the sciences, but we appeal to a climate science case study to illustrate what is at stake. Our strategy is to analyse the intuitive claims in (...) light of prominent accounts of confirmation of model predictions. We show that on the Bayesian account of confirmation, and also on the standard classical hypothesis-testing account, claims (a) and (b) are not generally true; but for some select cases, it is possible to distinguish data used for calibration from use-novel data, where only the latter confirm. The more specialized classical model-selection methods, on the other hand, uphold a nuanced version of claim (a), but this comes apart from (b), which must be rejected in favour of a more refined account of the relationship between calibration and confirmation. Thus, depending on the framework of confirmation, either the scope or the simplicity of the intuitive position must be revised. (shrink)

Bogen and Woodward’s distinction between data and phenomena raises the need to understand the structure of the data-to-phenomena and theory-to-phenomena inferences. I suggest that one way to study the structure of these inferences is to analyze the role of the assumptions involved in the inferences: What kind of assumptions are they? How do these assumptions contribute to the practice of identifying phenomena? In this paper, using examples from atmospheric dynamics, I develop an account of the practice of identifying the target (...) in the data-to-phenomena and theory-to-phenomena inferences in which assumptions about spatiotemporal scales play a central role in the identification of parameters that describe the target system. I also argue that these assumptions are not only empirical but they are also idealizing and abstracting. I conclude the paper with a reflection on the role of idealizations in modeling. (shrink)