North American Climate in CMIP5 Experiments. Part I: Evaluation of Historical Simulations of Continental and Regional Climatology

Abstract

This is the first part of a three-part paper on North American climate in phase 5 of the Coupled Model Intercomparison Project (CMIP5) that evaluates the historical simulations of continental and regional climatology with a focus on a core set of 17 models. The authors evaluate the models for a set of basic surface climate and hydrological variables and their extremes for the continent. This is supplemented by evaluations for selected regional climate processes relevant to North American climate, including cool season western Atlantic cyclones, the North American monsoon, the U.S. Great Plains low-level jet, and Arctic sea ice. In general, the multimodel ensemble mean represents the observed spatial patterns of basic climate and hydrological variables but with large variability across models and regions in the magnitude and sign of errors. No single model stands out as being particularly better or worse across all analyses, although some models consistently outperform the others for certain variables across most regions and seasons and higher-resolution models tend to perform better for regional processes. The CMIP5 multimodel ensemble shows a slight improvement relative to CMIP3 models in representing basic climate variables, in terms of the mean and spread, although performance has decreased for some models. Improvements in CMIP5 model performance are noticeable for some regional climate processes analyzed, such as the timing of the North American monsoon. The results of this paper have implications for the robustness of future projections of climate and its associated impacts, which are examined in the third part of the paper.

Evaluation of extreme rainfall and temperature over North America in CanRCM4 and CRCM5

“…However, while CanRCM4 simulates too little convective rainfall, largescale rainfall is over-estimated. The spatial extent of the warm bias in TXx by CanRCM4 is similar to the warm bias in mean temperature in CanESM2 (Sheffield et al 2013). Climate models from the same institution are known to often share the same biases (Knutti et al 2009).…”

“…4d, e). There is a warm mean temperature bias in CanESM2 in a similar region (Sheffield et al 2013). Although not as large as the warm bias in the Can-RCM4 simulations, CRCM5 is 2-3 °C warmer than observations in summer in the Central region.…”

Impact of anthropogenic climate change on wildfire across western US forests

“…S2-S4 and Table S2). The largest anthropogenic increases in standardized fuel aridity were present across the intermountain western United States, due in part to larger modeled warming rates relative to more maritime areas (27). Among aridity metrics, the largest increases tied to the ACC signal were for VPD and ETo because the interannual variability of these variables is primarily driven by temperature for much of the study area (28).…”

Projecting Future Change in Growing Degree Days for Winter Wheat

“…We used local scale projections from two GCMs: the Community Climate System Model 4 (CCSM4-realization r6i1p1, [31]) and the Model for Interdisciplinary Research on Climate 5 (MIROC5-realization r1i1p1, [32]). These two models were chosen because they performed appropriately over the South Central U.S. during the historical period and had relatively smaller cumulative biases ( Figure 2) [33]. The future projections used Representative Concentration Pathway (RCP) 8.5-a "business as usual" future trajectory for greenhouse gas concentrations and the RCP 4.5.…”