Evaluation of CMIP5 20^th century climate simulations for the Pacific Northwest USA

Abstract

[1] Monthly temperature and precipitation data from 41 global climate models (GCMs) of the Coupled Model Intercomparison Project Phase 5 (CMIP5) were compared to observations for the 20 th century, with a focus on the United States Pacific Northwest (PNW) and surrounding region. A suite of statistics, or metrics, was calculated, that included correlation and variance of mean seasonal spatial patterns, amplitude of seasonal cycle, diurnal temperature range, annual-to decadal-scale variance, long-term persistence, and regional teleconnections to El Niño Southern Oscillation (ENSO). Performance, or credibility, was assessed based on the GCMs' abilities to reproduce the observed metrics. GCMs were ranked in their credibility using two methods. The first simply treated all metrics equally. The second method considered two properties of the metrics: (1) redundancy of information (dependence) among metrics, and (2) confidence in the reliability of an individual metric for accurately ranking models. Confidence was related to how robust the estimate of the metric was to ensemble size, given that for most of the models only a small number of ensemble members (i.e., realizations of the 20 th century) were available. A cursory comparison with 24 CMIP3 models revealed few differences between the two generations of models with respect to the statistics analyzed.

Selecting climate change scenarios using impact‐relevant sensitivities

“…Pacific Northwest (PNW), USA, as in Rupp et al [2013]. This approach is computationally efficient and, by estimating one or two key impact variables, provides a richer framework than just ranges of GCM-derived changes in temperature (T ) and precipitation (P).…”

“…The ability of GCMs to reproduce observed climate can be quantified in many ways, and model ranking depends upon aspects of the climate analyzed and analysis methods [e.g., Brekke et al, 2008;Pierce et al, 2009;Rupp et al, 2013]. For each watershed and ecoregion, we used the evaluation by Rupp et al [2013], who ranked GCMs from CMIP5 based on numerous metrics of the twentieth century climate of the PNW and surrounding area.…”

“…For each watershed and ecoregion, we used the evaluation by Rupp et al [2013], who ranked GCMs from CMIP5 based on numerous metrics of the twentieth century climate of the PNW and surrounding area. Note that by ranking models by past performance, we assume that there is a connection between past and future performances-that is, in general, GCMs that better reproduce observed climate are likely to better simulate the future response to changing forcings.…”

“…Studies have offered guidance on GCM selection [Mote et al, 2011;Snover et al, 2013] and ways to rank GCM performance at the regional scale [Giorgi and Mearns, 2002;Brekke et al, 2008;Pitman and Perkins, 2008;Pierce et al, 2009;Mote and Salathé, 2010;Rupp et al, 2013] with the aim of selecting or weighting models. In this paper, building on such guidance, we demonstrate a technique that employs a quantitative, objective approach that aids selection, especially in situations when only a small number of GCMs can be selected.…”

Evaluation of CMIP5 Global Climate Models over the Volta Basin: Precipitation

“…ey also indicated the high performance of the multimodel ensemble average compared to the individual models. Other studies have evaluated the performance of CMIP5 models over other regions including North Pacific [21], Europe [22][23][24], and Asia (e.g., [25][26][27][28]) and reported high variations in the performances of the GCMs over the different regions.…”