Search

Global Climate Models are the main tools for climate projections. Since many models exist, it is common to use Multi-Model Ensembles to reduce biases and assess uncertainties in climate projections. Several approaches have been proposed to combine individual models and extract a robust signal from an ensemble. Among them, the Multi-Model Mean (MMM) is the most commonly used. Based on the assumption that the models are centered around the truth, it consists in averaging the ensemble, with the possibility of using equal weights for all models or to adjust weights to favor some models. In this paper, we propose a new alternative to reconstruct multi-decadal means of climate variables from a Multi-Model Ensemble, where the local performance of the models is taken into account. This is in contrast with MMM where a model has the same weight for all locations. Our approach is based on a computer vision method called graph cuts and consists in selecting for each grid point the most appropriate model, while at the same time considering the overall spatial consistency of the resulting field. The performance of the graph cuts approach is assessed based on two experiments: one where the ERA5 reanalyses are considered as the reference, and another involving a perfect model experiment where each model is in turn considered as the reference. We show that the graph cuts approach generally results in lower biases than other model combination approaches such as MMM, while at the same time preserving a similar level of spatial continuity. [ABSTRACT FROM AUTHOR]

Eastern China is regularly exposed to extreme precipitation with significant socio-economical consequences. Following an observational analysis in a first part of this study, here the ability of the Coupled Model Intercomparison Project Phase 6 models to reproduce the main modes of interannual variability of 5-day summer extreme precipitation over eastern China is evaluated, using an empirical orthogonal teleconnection (EOT) method. These models capture the main patterns and magnitudes of the different EOT patterns, although the North China mode is less well represented. Models also reproduce the dynamical features associated with each mode. There is no systematic improvement in the ability of models to simulate either the pattern or the 5-day intensity when using higher resolution models compared to coarser resolution ones. Instead, multi-member or multi-model ensembles lead to results closer to observations. Using a low mitigation projection pathway (SSP-370), it is shown that the risk of the most extreme 5-day precipitation events by about 40%, 80% and more than 150% for global-mean warming levels, relative to 1850–1900, of + 1.5, + 2 and + 3 ∘ C respectively. This increase is found to be more significant for 5-days events than for seasonal scale precipitation, consistent with previous studies. [ABSTRACT FROM AUTHOR]

Studies that evaluate climate change projections over the whole of South America (SA) and including different seasons and models are scarce. In this context, the objective of this work is to assess climate projections for SA through the use of climatic indices, considering the entire continent, distinct seasons, and ensembles of models. Projections performed with the Eta regional climate model and its driving global climate models (GCMs) are analysed. From these projections, 19 climate indices based on daily precipitation and maximum and minimum temperature are computed. The results focus on two ensembles (GCMs and Eta), time slices (1980–2005 and 2050–2080), and scenarios (RCP4.5 and RCP8.5). In the validation of the present climate, it is shown that Eta adds value to GCMs. Future projections indicate, for both austral summer (DJF) and winter (JJA), an increase in the frequency and intensity of extreme events of daily rainfall over southeastern and extreme north of SA. Over the Amazon, during DJF, there is a statistically significant increase in the number of consecutive dry days and a decrease in the consecutive wet days. For northeastern Brazil, these features are more intense in JJA. The frequency of cold (warm) nights and days is projected to decrease (increase) over the whole continent and seasons. The climate change signal for the 19 climate indices is more intense under RCP8.5, and the regions more vulnerable to climate change are the Amazon, northeastern Brazil, and southeastern SA. Considering Brazil, the projections of precipitation and air temperature are also shown by biomes. [ABSTRACT FROM AUTHOR]