Challenges in Combining Projections from Multiple Climate Models

By Reto Knutti, Reinhard Furrer, C. Tebaldi, Jan Cermak, Gerald A. Meehl. 2010

1,243 citations recorded by the source corpus.

Subfields: Atmospheric Science, Global and Planetary Change, Management Science and Operations Research

View Paper: https://www.semanticscholar.org/paper/d9f4c9f419772a290684da72c310a2b2d2e42ffa

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Atmospheric Science

In Global and Planetary Change

In Management Science and Operations Research

Citations in the corpus

Climate model dependence and the replicate Earth paradigm (2013, background): “…“best guess” 74 scenarios in the Intergovernmental Panel on Climate Change’s Fourth Assessment Report 75 (henceforth IPCC AR4; Meehl et al, 2007b), and was reinforced in the recent Expert Meeting on 76 Assessing and Combining Multi-model Climate Projections in 2010 (Knutti et al, 2010a).”
Climate Models, Calibration, and Confirmation (2013, neutral): “Note that Mayo’s ‘severe testing’ approach to confirmation does not support the Worrall conclusion either (see Mayo’s ([2010]) response to Worrall).”
Reflections Uncertainty and Decision Making in Climate Change Economics (2014, background): “All of these complications make it difficult to interpret the information in Figure 1 in a standard probabilistic framework without making some strong subjective judgements about how to combine the climate sensitivity estimates from the different studies (Knutti 2010; Knutti et al. 2010).”
Statistical problems in the probabilistic prediction of climate change (2012, methodology): “Because of these relationships between viable climate models, the outputs in MMEs are not independent as have been found empirically by noting that the sample variance of the MME does not decrease as the reciprocal of the ensemble size (Knutti et al., 2010b)”
Reliability of the CMIP3 ensemble (2010, background): “An immediate consequence of this is that, as models are added to an ensemble, the multi-model mean does not converge to observations as rapidly as would be expected for independent biases [Knutti et al., 2010].”
Exploring consensus in 21st century projections of climatically suitable areas for African vertebrates (2011, methodology): “…availability, and may thus not sample the full range of uncertainty or guarantee independence of models (Meehl et al. , 2007; Tebaldi & Knutti, 2007; Knutti et al. , 2008), but the large number of models included is expected to minimize potential biases by model choice (Knutti et al. , 2010).”
Climate model genealogy: Generation CMIP5 and how we got there (2013, methodology): “[3] Because formal methods to quantify uncertainties in projections are complex and direct observational constraints often absent [Knutti et al., 2010; Tebaldi and Knutti, 2007; Weigel et al., 2010], the spread of an ensemble of models is often used as a first-order estimate of projection uncertaint”
An asynchronous regional regression model for statistical downscaling of daily climate variables (2013, methodology): “Smoothing AOGCM output has been previously recommended (e.g. Raisanen and Ylhaisi, 2011), and it has been suggested that the smoothing that results from averaging may be one of the reasons why ensemble AOGCM projections typically outperform any individual model simulation (Knutti et al., 2010).”
Reduced global warming from CMIP6 projections when weighting models by performance and independence (2020, background): “Thus, when calculating probabilities based on such MMEs it is important to account for model interdependence in order to accurately translate model spread into estimates of mean change and related uncertainties (Knutti, 2010; Knutti et al., 2010).”
Skill and independence weighting for multi-model assessments (2016, background): “drying in the future (just as each individual model exhibits some regions of 406 extreme drying, but the lack of agreement amongst models on where those 407 regions are causes the multi-model mean to lack any such behavior, as noted in 408 Knutti et al (2010) [26]).”