Can multi‐model combination really enhance the prediction skill of probabilistic ensemble forecasts?

By A. Weigel, Markus Liniger, C. Appenzeller. 2007

349 citations recorded by the source corpus.

Subfields: Atmospheric Science, Global and Planetary Change

View Paper: https://www.semanticscholar.org/paper/d1dbc822a606e99cdfcb4d71f900056abfdd216b

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

Citations in the corpus

Forecast combinations: An over 50-year review (2022, background): “In this light, two major sources of forecast uncertainty, initial condition uncertainty resulting from the chaotic nature of the atmosphere, and model uncertainty arising from imperfect numerical models, are addressed (Lorenz, 1963; Weigel et al., 2008; Baran, 2014).”
Assessment of CMIP5 climate models and projected temperature changes over Northern Eurasia (2014, background): “Because single models are overconfident (Weigel et al 2008) and multi-model ensembles contain information from all participating models (Pincus et al 2008), it is generally believed that multi-model ensembles are superior to single models (IPCC 2001, Duan and Phillips 2010, Miao et al 2013).”
Bayesian estimation of local signal and noise in multimodel simulations of climate change (2010, background): “A possible explanation is that the multimodel ensemble embraces distinctly different physical parameterizations, thus surmounting the limitations of an overconfident single‐ model simulation [Tebaldi and Knutti, 2007; Pincus et al., 2008; Weigel et al., 2008 ].”
The end of model democracy? (2010, background): “Surprisingly, even adding a poor model can improve a prediction if the individual models tend to be overconfident (Weigel et al. 2008).”
Challenges in Combining Projections from Multiple Climate Models (2010, background): “Weighted averages are found to perform better in many cases (Robertson et al. 2004; Min and Hense 2006; Peña and Van den Dool 2008; Weigel et al. 2008), provided that sufficient information is available to determine the weight (see section 2c).”
Contrasted impacts of climate change on stream fish assemblages along an environmental gradient (2009, background): “Although very common in various ﬁelds such as meteorology (e.g. Weigel et al. , 2008) or economics (e.g. Gregory et al. , 2001), these consensus methods and ensemble forecasting approaches have been only recently applied in ecology (Thuiller, 2004; Araujo et al. , 2005, 2006; Araujo & New, 2007;…”
Assessment of representations of model uncertainty in monthly and seasonal forecast ensembles (2011, background): “[17] As has been demonstrated in numerous studies [e.g., Palmer et al., 2004; Weigel et al., 2008], the multi‐model combination effectively reduces the RMSE while the ensemble spread is increased leading to overall improved skill.”
Trends of Climate Change in Saudi Arabia: Implications on Water Resources (2018, positive): “The multi-model ensemble combination showed an enhanced predictive capacity for climatic parameters in comparison to a single model [8].”
Seasonal climate forecasting (2010, background): “Weigel et al. (2008) provided one of the most satisfactory explanations to date.”
Predicting Sudden Stratospheric Warming 2018 and Its Climate Impacts With a Multimodel Ensemble (2018, background): “In our approach, multimodel ensembles are formed from individual ensemble members, and therefore, different models have different weight in multimodel mean (MMM) proportional to their number of ensemble members (the pooled ensemble method is discussed by Weigel et al., 2008).”
Medium‐range multimodel ensemble combination and calibration (2009, background): “It has been shown in the context of seasonal forecasting (Palmer et al., 2004; Hagedorn et al., 2005; Weigel et al., 2008) that the combination of ensembles from different models results in more skill than the single ensembles considered separately, this improvement being not just due to the…”
ENSEMBLES: A new multi‐model ensemble for seasonal‐to‐annual predictions—Skill and progress beyond DEMETER in forecasting tropical Pacific SSTs (2009, background): “The central reason for enhanced skill in a MME is the reduction of the overconfidence of the single-model ensembles, i.e., transforming the largely under-dispersive forecasts from each single-model ensemble into a better-dispersed MME [Weigel et al., 2008].”