Uncertainty in climate change projections: the role of internal variability

Uncertainty in future climate change presents a key challenge for adaptation planning. In this study, uncertainty arising from internal climate variability is investigated using a new 40-member ensemble conducted with the National Center for Atmospheric Research Community Climate System Model Version 3 (CCSM3) under the SRES A1B greenhouse gas and ozone recovery forcing scenarios during 2000–2060. The contribution of intrinsic atmospheric variability to the total uncertainty is further examined using a 10,000-year control integration of the atmospheric model component of CCSM3 under fixed boundary conditions. The global climate response is characterized in terms of air temperature, precipitation, and sea level pressure during winter and summer. The dominant source of uncertainty in the simulated climate response at middle and high latitudes is internal atmospheric variability associated with the annular modes of circulation variability. Coupled ocean-atmosphere variability plays a dominant role in the tropics, with attendant effects at higher latitudes via atmospheric teleconnections. Uncertainties in the forced response are generally larger for sea level pressure than precipitation, and smallest for air temperature. Accordingly, forced changes in air temperature can be detected earlier and with fewer ensemble members than those in atmospheric circulation and precipitation. Implications of the results for detection and attribution of observed climate change and for multi-model climate assessments are discussed. Internal variability is estimated to account for at least half of the inter-model spread in projected climate trends during 2005–2060 in the CMIP3 multi-model ensemble.     

Scale-dependent regional climate predictability over North America inferred from CMIP3 and CMIP5 ensemble simulations

Through the analysis of ensembles of coupled model simulations and projections collected from CMIP3 and CMIP5, we demonstrate that a fundamental spatial scale limit might exist below which useful additional refinement of climate model predictions and projections may not be possible. That limit varies among climate variables and from region to region. We show that the uncertainty (noise) in surface temperature predictions (represented by the spread among an ensemble of global climate model simulations) generally exceeds the ensemble mean (signal) at horizontal scales below 1000 km throughout North America, implying poor predictability at those scales. More limited skill is shown for the predictability of regional precipitation. The ensemble spread in this case tends to exceed or equal the ensemble mean for scales below 2000 km. These findings highlight the challenges in predicting regionally specific future climate anomalies, especially for hydroclimatic impacts such as drought and wetness.     

Transient climate changes in a perturbed parameter ensemble of emissions-driven earth system model simulations

We describe results from a 57-member ensemble of transient climate change simulations, featuring simultaneous perturbations to 54 parameters in the atmosphere, ocean, sulphur cycle and terrestrial ecosystem components of an earth system model (ESM). These emissions-driven simulations are compared against the CMIP3 multi-model ensemble of physical climate system models, used extensively to inform previous assessments of regional climate change, and also against emissions-driven simulations from ESMs contributed to the CMIP5 archive. Members of our earth system perturbed parameter ensemble (ESPPE) are competitive with CMIP3 and CMIP5 models in their simulations of historical climate. In particular, they perform reasonably well in comparison with HadGEM2-ES, a more sophisticated and expensive earth system model contributed to CMIP5. The ESPPE therefore provides a computationally cost-effective tool to explore interactions between earth system processes. In response to a non-intervention emissions scenario, the ESPPE simulates distributions of future regional temperature change characterised by wide ranges, and warm shifts, compared to those of CMIP3 models. These differences partly reflect the uncertain influence of global carbon cycle feedbacks in the ESPPE. In addition, the regional effects of interactions between different earth system feedbacks, particularly involving physical and ecosystem processes, shift and widen the ESPPE spread in normalised patterns of surface temperature and precipitation change in many regions. Significant differences from CMIP3 also arise from the use of parametric perturbations (rather than a multimodel ensemble) to represent model uncertainties, and this is also the case when ESPPE results are compared against parallel emissions-driven simulations from CMIP5 ESMs. When driven by an aggressive mitigation scenario, the ESPPE and HadGEM2-ES reveal significant but uncertain impacts in limiting temperature increases during the second half of the twenty-first century. Emissions-driven simulations create scope for development of errors in properties that were previously prescribed in coupled ocean–atmosphere models, such as historical CO₂ concentrations and vegetation distributions. In this context, historical intra-ensemble variations in the airborne fraction of CO₂ emissions, and in summer soil moisture in northern hemisphere continental regions, are shown to be potentially useful constraints, subject to uncertainties in the relevant observations. Our results suggest that future climate-related risks can be assessed more comprehensively by updating projection methodologies to support formal combination of emissions-driven perturbed parameter and multi-model earth system model simulations with suitable observational constraints. This would provide scenarios underpinned by a more complete representation of the chain of uncertainties from anthropogenic emissions to future climate outcomes.     

A multi-physics ensemble of regional climate change projections over the Iberian Peninsula

This study illustrates the sensitivity of regional climate change projections to the model physics. A single-model (MM5) multi-physics ensemble of regional climate simulations over the Iberian Peninsula for present (1970–1999) and future (2070–2099 under the A2 scenario) periods is assessed. The ensemble comprises eight members resulting from the combination of two options of parameterization schemes for the planetary boundary layer, cumulus and microphysics. All the considered combinations were previously evaluated by comparing hindcasted simulations to observations, none of them providing clearly outlying climates. Thus, the differences among the various ensemble members (spread) in the future projections could be considered as a matter of uncertainty in the change signals (as similarly assumed in multi-model studies). The results highlight the great dependence of the spread on the synoptic conditions driving the regional model. In particular, the spread generally amplifies under the future scenario leading to a large spread accompanying the mean change signals, as large as the magnitude of the mean projected changes and analogous to the spread obtained in multi-model ensembles. Moreover, the sign of the projected change varies depending on the choice of the model physics in many cases. This, together with the fact that the key mechanisms identified for the simulation of the climatology of a given period (either present or future) and those introducing the largest spread in the projected changes differ significantly, make further claims for efforts to better understand and model the parameterized subgrid processes.     

Regional and seasonal intercomparison of CMIP3 and CMIP5 climate model ensembles for temperature and precipitation

Regional and seasonal temperature and precipitation over land are compared across two generations of global climate model ensembles, specifically, CMIP5 and CMIP3, through historical twentieth century skills and multi-model agreement, and twenty first century projections. A suite of diagnostic and performance metrics, ranging from spatial bias or model-consensus maps and aggregate time series plots, to measures of equivalence between probability density functions and Taylor diagrams, are used for the intercomparisons. Pairwise and multi-model ensemble comparisons were performed for 11 models, which were selected based on data availability and resolutions. Results suggest little change in the central tendency or variability or uncertainty of historical skills or consensus across the two generations of models. However, there are regions and seasons, at different levels of aggregation, where significant changes, performance improvements, and even degradation in skills, are suggested. The insights may provide directions for further improvements in next generations of climate models, and in the meantime, help inform adaptation and policy.     

On the generation of climate model ensembles

Climate model ensembles are used to estimate uncertainty in future projections, typically by interpreting the ensemble distribution for a particular variable probabilistically. There are, however, different ways to produce climate model ensembles that yield different results, and therefore different probabilities for a future change in a variable. Perhaps equally importantly, there are different approaches to interpreting the ensemble distribution that lead to different conclusions. Here we use a reduced-resolution climate system model to compare three common ways to generate ensembles: initial conditions perturbation, physical parameter perturbation, and structural changes. Despite these three approaches conceptually representing very different categories of uncertainty within a modelling system, when comparing simulations to observations of surface air temperature they can be very difficult to separate. Using the twentieth century CMIP5 ensemble for comparison, we show that initial conditions ensembles, in theory representing internal variability, significantly underestimate observed variance. Structural ensembles, perhaps less surprisingly, exhibit over-dispersion in simulated variance. We argue that future climate model ensembles may need to include parameter or structural perturbation members in addition to perturbed initial conditions members to ensure that they sample uncertainty due to internal variability more completely. We note that where ensembles are over- or under-dispersive, such as for the CMIP5 ensemble, estimates of uncertainty need to be treated with care.     

Identifying uncertainties in Arctic climate change projections

Wide ranging climate changes are expected in the Arctic by the end of the 21st century, but projections of the size of these changes vary widely across current global climate models. This variation represents a large source of uncertainty in our understanding of the evolution of Arctic climate. Here we systematically quantify and assess the model uncertainty in Arctic climate changes in two CO₂ doubling experiments: a multimodel ensemble (CMIP3) and an ensemble constructed using a single model (HadCM3) with multiple parameter perturbations (THC-QUMP). These two ensembles allow us to assess the contribution that both structural and parameter variations across models make to the total uncertainty and to begin to attribute sources of uncertainty in projected changes. We find that parameter uncertainty is an major source of uncertainty in certain aspects of Arctic climate. But also that uncertainties in the mean climate state in the 20th century, most notably in the northward Atlantic ocean heat transport and Arctic sea ice volume, are a significant source of uncertainty for projections of future Arctic change. We suggest that better observational constraints on these quantities will lead to significant improvements in the precision of projections of future Arctic climate change.     

Reliability and importance of structural diversity of climate model ensembles

We investigate the performance of the newest generation multi-model ensemble (MME) from the Coupled Model Intercomparison Project (CMIP5). We compare the ensemble to the previous generation models (CMIP3) as well as several single model ensembles (SMEs), which are constructed by varying components of single models. These SMEs range from ensembles where parameter uncertainties are sampled (perturbed physics ensembles) through to an ensemble where a number of the physical schemes are switched (multi-physics ensemble). We focus on assessing reliability against present-day climatology with rank histograms, but also investigate the effective degrees of freedom (EDoF) of the fields of variables which makes the statistical test of reliability more rigorous, and consider the distances between the observation and ensemble members. We find that the features of the CMIP5 rank histograms, of general reliability on broad scales, are consistent with those of CMIP3, suggesting a similar level of performance for present-day climatology. The spread of MMEs tends towards being “over-dispersed” rather than “under-dispersed”. In general, the SMEs examined tend towards insufficient dispersion and the rank histogram analysis identifies them as being statistically distinguishable from many of the observations. The EDoFs of the MMEs are generally greater than those of SMEs, suggesting that structural changes lead to a characteristically richer range of model behaviours than is obtained with parametric/physical-scheme-switching ensembles. For distance measures, the observations and models ensemble members are similarly spaced from each other for MMEs, whereas for the SMEs, the observations are generally well outside the ensemble. We suggest that multi-model ensembles should represent an important component of uncertainty analysis.     

The first multi-model ensemble of regional climate simulations at kilometer-scale resolution part 2: historical and future simulations of precipitation

This paper presents the first multi-model ensemble of 10-year, “convection-permitting” kilometer-scale regional climate model (RCM) scenario simulations downscaled from selected CMIP5 GCM projections for historical and end of century time slices. The technique is to first downscale the CMIP5 GCM projections to an intermediate 12–15 km resolution grid using RCMs, and then use these fields to downscale further to the kilometer scale. The aim of the paper is to provide an overview of the representation of the precipitation characteristics and their projected changes over the greater Alpine domain within a Coordinated Regional Climate Downscaling Experiment Flagship Pilot Study and the European Climate Prediction system project, tasked with investigating convective processes at the kilometer scale. An ensemble of 12 simulations performed by different research groups around Europe is analyzed. The simulations are evaluated through comparison with high resolution observations while the complementary ensemble of 12 km resolution driving models is used as a benchmark to evaluate the added value of the convection-permitting ensemble. The results show that the kilometer-scale ensemble is able to improve the representation of fine scale details of mean daily, wet-day/hour frequency, wet-day/hour intensity and heavy precipitation on a seasonal scale, reducing uncertainty over some regions. It also improves the representation of the summer diurnal cycle, showing more realistic onset and peak of convection. The kilometer-scale ensemble refines and enhances the projected patterns of change from the coarser resolution simulations and even modifies the sign of the precipitation intensity change and heavy precipitation over some regions. The convection permitting simulations also show larger changes for all indices over the diurnal cycle, also suggesting a change in the duration of convection over some regions. A larger positive change of frequency of heavy to severe precipitation is found. The results are encouraging towards the use of convection-permitting model ensembles to produce robust assessments of the local impacts of future climate change.

     

Climate model errors, feedbacks and forcings: a comparison of perturbed physics and multi-model ensembles

Ensembles of climate model simulations are required for input into probabilistic assessments of the risk of future climate change in which uncertainties are quantified. Here we document and compare aspects of climate model ensembles from the multi-model archive and from perturbed physics ensembles generated using the third version of the Hadley Centre climate model (HadCM3). Model-error characteristics derived from time-averaged two-dimensional fields of observed climate variables indicate that the perturbed physics approach is capable of sampling a relatively wide range of different mean climate states, consistent with simple estimates of observational uncertainty and comparable to the range of mean states sampled by the multi-model ensemble. The perturbed physics approach is also capable of sampling a relatively wide range of climate forcings and climate feedbacks under enhanced levels of greenhouse gases, again comparable with the multi-model ensemble. By examining correlations between global time-averaged measures of model error and global measures of climate change feedback strengths, we conclude that there are no simple emergent relationships between climate model errors and the magnitude of future global temperature change. Algorithms for quantifying uncertainty require the use of complex multivariate metrics for constraining projections.     

Uncertainties in regional climate change prediction: a regional analysis of ensemble simulations with the HADCM2 coupled AOGCM

We analyze ensembles (four realizations) of historical and future climate transient experiments carried out with the coupled atmosphere-ocean general circulation model (AOGCM) of the Hadley Centre for Climate Prediction and Research, version HADCM2, with four scenarios of greenhouse gas (GHG) and sulfate forcing. The analysis focuses on the regional scale, and in particular on 21 regions covering all land areas in the World (except Antarctica). We examine seasonally averaged surface air temperature and precipitation for the historical period of 1961–1990 and the future climate period of 2046–2075. Compared to previous AOGCM simulations, the HADCM2 model shows a good performance in reproducing observed regional averages of summer and winter temperature and precipitation. The model, however, does not reproduce well observed interannual variability. We find that the uncertainty in regional climate change predictions associated with the spread of different realizations in an ensemble (i.e. the uncertainty related to the internal model variability) is relatively low for all scenarios and regions. In particular, this uncertainty is lower than the uncertainty due to inter-scenario variability and (by comparison with previous regional analyses of AOGCMs) with inter-model variability. The climate biases and sensitivities found for different realizations of the same ensemble were similar to the corresponding ensemble averages and the averages associated with individual realizations of the same ensemble did not differ from each other at the 5% confidence level in the vast majority of cases. These results indicate that a relatively small number of realizations (3 or 4) is sufficient to characterize an AOGCM transient climate change prediction at the regional scale. Received: 12 January 1998 / Accepted: 7 July 1999     

Extended regional climate model projections for Europe until the mid-twentyfirst century: combining ENSEMBLES and CMIP3

This study aims at sharpening the existing knowledge of expected seasonal mean climate change and its uncertainty over Europe for the two key climate variables air temperature and precipitation amount until the mid-twentyfirst century. For this purpose, we assess and compensate the global climate model (GCM) sampling bias of the ENSEMBLES regional climate model (RCM) projections by combining them with the full set of the CMIP3 GCM ensemble. We first apply a cross-validation in order to assess the skill of different statistical data reconstruction methods in reproducing ensemble mean and standard deviation. We then select the most appropriate reconstruction method in order to fill the missing values of the ENSEMBLES simulation matrix and further extend the matrix by all available CMIP3 GCM simulations forced by the A1B emission scenario. Cross-validation identifies a randomized scaling approach as superior in reconstructing the ensemble spread. Errors in ensemble mean and standard deviation are mostly less than 0.1 K and 1.0 % for air temperature and precipitation amount, respectively. Reconstruction of the missing values reveals that expected seasonal mean climate change of the ENSEMBLES RCM projections is not significantly biased and that the associated uncertainty is not underestimated due to sampling of only a few driving GCMs. In contrast, the spread of the extended simulation matrix is partly significantly lower, sharpening our knowledge about future climate change over Europe by reducing uncertainty in some regions. Furthermore, this study gives substantial weight to recent climate change impact studies based on the ENSEMBLES projections, since it confirms the robustness of the climate forcing of these studies concerning GCM sampling.     

The impact of perturbations to ocean-model parameters on climate and climate change in a coupled model

Assessments of the impacts of uncertainties in parameters on mean climate and climate change in complex climate models have, to date, largely focussed on perturbations to parameters in the atmosphere component of the model. Here we expand on a previously published study which found the global impacts of perturbed ocean parameters on the rate of transient climate change to be small compared to perturbed atmosphere parameters. By separating the climate-change-induced ocean vertical heat transport in each perturbed member into components associated with the resolved flow and each parameterisation scheme, we show that variations in global mean heat uptake in different perturbed versions are an order of magnitude smaller than the average heat uptake. The lack of impact of the perturbations is attributed to (1) the relatively small impact of the perturbation on the direct vertical heat transport associated with the perturbed process and (2) a compensation between those direct changes and indirect changes in heat transport from other processes. Interactions between processes and changes appear to combine in complex ways to limit ensemble spread and uncertainty in the rate of warming. We also investigate regional impacts of the perturbations that may be important for climate change predictions. We find variations across the ensemble that are significant when measured against natural variability. In terms of the experimental set-up used here (models without flux adjustments) we conclude that perturbed physics ensembles with ocean parameter perturbations are an important component of any probabilistic estimate of future climate change, despite the low spread in global mean quantities. Hence, careful consideration should be given to assessing uncertainty in ocean processes in future probabilistic assessments of regional climate change.     

CMIP5多模式对东亚地区地面气温年际变率的回报研究

对CMIP5全球气候模式中年代际回报试验的气温资料及其简单集合平均(Multi-model ensemble mean,EMN)和贝叶斯模式平均的结果(Bayesian Model Averaging,BMA)进行经验正交函数(Empirical Orthogonal Function,EOF)分解和Morlet小波分析,检验评估各个模式及其EMN和BMA对东亚地面气温的方差、气温时空分布特征及周期变化的回报能力。结果表明,10个模式、EMN、BMA都能很好地回报出1981—2010年东亚地面气温的方差分布,其中BMA回报效果最好。EOF分析表明,BMA能较好地回报出东亚地面气温第一模态的时空分布。MIROC5能较好地回报出第二模态的趋势变化,但却不能回报出气温的年际变率。绝大多数模式和EMN、BMA虽然能回报出东亚地面气温的变化趋势,但是对气温年际变率的回报仍然是比较困难的。CMCC-CM对气温变化主模态的3~5 a的周期变化特征回报效果最好,和NCEP资料的结果最为接近。     

Assessing the robustness of projected precipitation changes over central Africa on the basis of a multitude of global and regional climate projections

It is well accepted within the scientific community that a large ensemble of different projections is required to achieve robust climate change information for a specific region. For this purpose we have compiled a state-of-the-art multi-model multi-scenario ensemble of global and regional precipitation projections. This ensemble combines several global projections from the CMIP3 and CMIP5 databases, along with some recently downscaled regional CORDEX-Africa projections. Altogether daily precipitation data from 77 different climate change projections is analysed; separated into 31 projections for a high and 46 for a low emission scenario. We find a robust indication that, independent of the underlying emission scenario, annual total precipitation amounts over the central African region are not likely to change severely in the future. However some robust changes in precipitation characteristics, like the intensification of heavy rainfall events as well as an increase in the number of dry spells during the rainy season are projected for the future. Further analysis shows that over some regions the results of the climate change assessment clearly depend on the size of the analyzed ensemble. This indicates the need of a “large-enough” ensemble of independent climate projections to allow for a reliable climate change assessment.     

Uncertainty estimation of the global temperature trends for multiple radiosondes, reanalyses, and CMIP3/IPCC climate model simulations

Based on three groups of datasets that include radiosondes, reanalyses, and climate model simulations (e.g., Coupled Model Intercomparison Project, CMIP3) from 1979 to 2008, the interannual variability, global temperature trends, and their uncertainty using ensemble spread among intra-group and inter-group datasets have been discussed. The results show that the interannual temperature variability increased from the troposphere to stratosphere, and the maximum occurs around 50?hPa. The CMIP3 climate models have the largest discrepancy in the stratosphere. The intra-group correlations at 500?hPa generally show high similarity within each data group while the inter-group correlations between reanalyses and the CMIP3 climate model simulations indicate lesser similarity. In contrast, the inter-group correlation at 50?hPa is improved except with the Japanese 25-year Reanalysis Project (JRA-25) dataset, and the Twentieth Century Reanalysis (20CR) reanalysis shows a weak cross correlation. The global temperature trends are highly dependent on the individual data sources. Compared to the radiosondes, the reanalyses show a large ensemble spread of trends in the stratosphere, and the CMIP3 climate model simulations have a large ensemble spread in the height of the crossover point where tropospheric warming changes into stratospheric cooling. The largest ensemble spread among the reanalyses in the stratosphere is mainly from the large discrepancy in the JRA-25 reanalysis after 1998 and a relatively weak anomaly in the 20CR before 1986. The largest ensemble spread among the CMIP3 climate models in the troposphere is related to the influence of both volcanic eruptions and El Ni?o/La Ni?a–Southern Oscillation events. The strong anomalies corresponding to the volcanic eruptions of El Chichon in 1982 and Mt Pinatubo in 1991 are clearly identified in the stratosphere. These volcanic eruptions reduced the warming in the troposphere and strengthened the cooling in the stratosphere during the most recent 30?years.     

Near-Term Projections of Global and Regional Land Mean Temperature Changes Considering Both the Secular Trend and Multidecadal Variability

Near-term climate projections are needed by policymakers; however, these projections are difficult because internally generated climate variations need to be considered. In this study, temperature change scenarios in the near-term period 2017–35 are projected at global and regional scales based on a refined multi-model ensemble approach that considers both the secular trend (ST) and multidecadal variability (MDV) in the Coupled Model Intercomparison Project Phase 5 (CMIP5) simulations. The ST and MDV components are adaptively extracted from each model simulation by using the ensemble empirical mode decomposition (EEMD) filter, reconstructed via the Bayesian model averaging (BMA) method for the historical period 1901–2005, and validated for 2006–16. In the simulations of the “medium” representative concentration pathways scenario during 2017–35, the MDV-modulated temperature change projected via the refined approach displays an increase of 0.44°C (90% uncertainty range from 0.30 to 0.58°C) for global land, 0.48°C (90% uncertainty range from 0.29 to 0.67°C) for the Northern Hemispheric land (NL), and 0.29°C (90% uncertainty range from 0.23 to 0.35°C) for the Southern Hemispheric land (SL). These increases are smaller than those projected by the conventional arithmetic mean approach. The MDV enhances the ST in 13 of 21 regions across the world. The largest MDV-modulated warming effect (46%) exists in central America. In contrast, the MDV counteracts the ST in NL, SL, and eight other regions, with the largest cooling effect (220%) in Alaska.     

Higher precision estimates of regional polar warming by ensemble regression of climate model projections

This study presents projections of twenty-first century wintertime surface temperature changes over the high-latitude regions based on the third Coupled Model Inter-comparison Project (CMIP3) multi-model ensemble. The state-dependence of the climate change response on the present day mean state is captured using a simple yet robust ensemble linear regression model. The ensemble regression approach gives different and more precise estimated mean responses compared to the ensemble mean approach. Over the Arctic in January, ensemble regression gives less warming than the ensemble mean along the boundary between sea ice and open ocean (sea ice edge). Most notably, the results show 3?°C less warming over the Barents Sea (~7?°C compared to ~10?°C). In addition, the ensemble regression method gives projections that are 30?% more precise over the Sea of Okhostk, Bering Sea and Labrador Sea. For the Antarctic in winter (July) the ensemble regression method gives 2?°C more warming over the Southern Ocean close to the Greenwich Meridian (~7?°C compared to ~5?°C). Projection uncertainty was almost half that of the ensemble mean uncertainty over the Southern Ocean between 30° W to 90° E and 30?% less over the northern Antarctic Peninsula. The ensemble regression model avoids the need for explicit ad hoc weighting of models and exploits the whole ensemble to objectively identify overly influential outlier models. Bootstrap resampling shows that maximum precision over the Southern Ocean can be obtained with ensembles having as few as only six climate models.     

Climate model dependence and the replicate Earth paradigm

Multi-model ensembles are commonly used in climate prediction to create a set of independent estimates, and so better gauge the likelihood of particular outcomes and better quantify prediction uncertainty. Yet researchers share literature, datasets and model code—to what extent do different simulations constitute independent estimates? What is the relationship between model performance and independence? We show that error correlation provides a natural empirical basis for defining model dependence and derive a weighting strategy that accounts for dependence in experiments where the multi-model mean would otherwise be used. We introduce the “replicate Earth” ensemble interpretation framework, based on theoretically derived statistical relationships between ensembles of perfect models (replicate Earths) and observations. We transform an ensemble of (imperfect) climate projections into an ensemble whose mean and variance have the same statistical relationship to observations as an ensemble of replicate Earths. The approach can be used with multi-model ensembles that have varying numbers of simulations from different models, accounting for model dependence. We use HadCRUT3 data and the CMIP3 models to show that in out of sample tests, the transformed ensemble has an ensemble mean with significantly lower error and much flatter rank frequency histograms than the original ensemble.     

Climate of Russia in the 21st Century. Part 2. Verification of atmosphere-ocean general circulation models CMIP3 for projections of future climate changes

Results of simulation of radiation, cloud cover, surface air temperature, sea-level pressure, and hydrological regime components for Russia with the help of an ensemble of CMIP3 global climate models is analyzed. Despite a large spread among the models, the CMIP3 AOGCM ensemble simulations of the key characteristics of the observed surface climate agree well with observations, anyway in averaging over areas of vast regions, from watersheds of large rivers to the whole of Russia. These means (ensemble-and area-averaged values) often fall into the range of estimates derived from observations. This suggests the existence of uncertainty in the estimates obtained from simulations as well as from observational data. Comparison of different-generation models demonstrates a gradual improvement of the AOGCM simulation of surface climate characteristics. In general, the averaging over the CMIP3 AOGCM ensemble allows us to state that the ensemble is suitable for estimates of future climate changes.