Holism, entrenchment, and the future of climate model pluralism

Lenhard, Johannes; Winsberg, Eric

doi:10.1016/j.shpsb.2010.07.001

Cited by 119 publications

(78 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Mauritsen et al 2012, 14). Such interdependencies among errors are often fully expected, even though exactly what the dependencies are is often unclear, and so earlier layers of model development are left intact (see also Lenhard and Winsberg 2010).…”

Section: Models Priorities and Valuesmentioning

confidence: 99%

Values and evidence: how models make a difference

Parker

Winsberg

2017

Euro Jnl Phil Sci

Self Cite

View full text Add to dashboard Cite

We call attention to an underappreciated way in which non-epistemic values influence evidence evaluation in science. Our argument draws upon some well-known features of scientific modeling. We show that, when scientific models stand in for background knowledge in Bayesian and other probabilistic methods for evidence evaluation, conclusions can be influenced by the non-epistemic values that shaped the setting of priorities in model development. Moreover, it is often infeasible to correct for this influence. We further suggest that, while this value influence is not particularly prone to the problem of wishful thinking, it could have problematic non-epistemic consequences in some cases.

show abstract

Section: Models Priorities and Valuesmentioning

confidence: 99%

Values and evidence: how models make a difference

Parker

Winsberg

2017

Euro Jnl Phil Sci

Self Cite

View full text Add to dashboard Cite

show abstract

“…These initial conditions and drivers have the role of auxiliary hypotheses in the Duhem/Quine sense (Stanford 2013) since they are themselves uncertain approximations to the real world conditions. Evaluation of a simulation (or often, an ensemble of simulations) is therefore a test of all the components (model physics, parameterisations, and auxiliaries) at once (Lenhard and Winsberg 2010). …”

mentioning

confidence: 99%

A practical philosophy of complex climate modelling

Schmidt

Sherwood

2014

Euro Jnl Phil Sci

View full text Add to dashboard Cite

We give an overview of the practice of developing and using complex climate models, as seen from experiences in a major climate modelling center and through participation in the Coupled Model Intercomparison Project (CMIP). We discuss the construction and calibration of models; their evaluation, especially through use of out-of-sample tests; and their exploitation in multi-model ensembles to identify biases and make predictions. We stress that adequacy or utility of climate models is best assessed via their skill against more naïve predictions. The framework we use for making inferences about reality using simulations is naturally Bayesian (in an informal sense), and has many points of contact with more familiar examples of scientific epistemology. While the use of complex simulations in science is a development that changes much in how science is done in practice, we argue that the concepts being applied fit very much into traditional practices of the scientific method, albeit those more often associated with laboratory work.

show abstract

“…While some descriptions of software design are starting to appear (e.g., Drake, 2005), detailed analysis of the design of global climate models remains a challenge because the models have undergone continued code modification for years, and, in some cases, decades. This makes a reductionist analysis of specific design decisions impossible, because each design decision is "generatively entrenched" (Lenhard and Winsberg, 2010) -that is, design features form a complex web because each has played a role in generating the others. Furthermore, each lab retains a deep but tacit knowledge base about their own models, which is readily apparent to anyone (Easterbrook and Johns, 2009), but hard to share through model intercomparison projects.…”

Section: Introductionmentioning

confidence: 99%

The software architecture of climate models: a graphical comparison of CMIP5 and EMICAR5 configurations

Alexander

Easterbrook

2015

Geosci. Model Dev.

View full text Add to dashboard Cite

Abstract. We analyze the source code of eight coupled climate models, selected from those that participated in the CMIP5 (Taylor et al., 2012) or EMICAR5 Zickfeld et al., 2013) intercomparison projects. For each model, we sort the preprocessed code into components and subcomponents based on dependency structure. We then create software architecture diagrams that show the relative sizes of these components/subcomponents and the flow of data between them. The diagrams also illustrate several major classes of climate model design; the distribution of complexity between components, which depends on historical development paths as well as the conscious goals of each institution; and the sharing of components between different modeling groups. These diagrams offer insights into the similarities and differences in structure between climate models, and have the potential to be useful tools for communication between scientists, scientific institutions, and the public.

show abstract

Holism, entrenchment, and the future of climate model pluralism

Cited by 119 publications

References 29 publications

Values and evidence: how models make a difference

Values and evidence: how models make a difference

A practical philosophy of complex climate modelling

The software architecture of climate models: a graphical comparison of CMIP5 and EMICAR5 configurations

Contact Info

Product

Resources

About