Search

Reconstructing past climates remains a difficult task because pre‐instrumental observational networks are composed of geographically sparse and noisy paleoclimate proxy records that require statistical techniques to inform complete climate fields. Traditionally, instrumental or climate model statistical relationships are used to spread information from proxy measurements to other locations and to other climate variables. Here ensembles drawn from single climate models and from combinations of multiple climate models are used to reconstruct temperature variability over the last millennium in idealized experiments. We find that reconstructions derived from multi‐model ensembles produce lower error than reconstructions from single‐model ensembles when reconstructing independent model and instrumental data. Specifically, we find the largest decreases in error over regions far from proxy locations that are often associated with large uncertainties in model physics, such as mid‐ and high‐latitude ocean and sea‐ice regions. Furthermore, we find that multi‐model ensemble reconstructions outperform single‐model reconstructions that use covariance localization. We propose that multi‐model ensembles could be used to improve paleoclimate reconstructions in time periods beyond the last millennium and for climate variables other than air temperature, such as drought metrics or sea ice variables. Plain Language Summary: Understanding past climate variability is important for contextualizing climate change as well as for testing the ability of climate models to simulate the climate system before global warming. However, reconstructing past climate variability remains a complex task because pre‐instrumental paleoclimate proxy records, such as tree rings, corals, ice cores, and sediment cores, are geographically sparse and are not perfect recorders of climate information. Exactly how to extrapolate information from paleoclimate proxies to other locations and climate variables remains an outstanding issue. Traditionally, information about how one location varies with another location or variable (covariance) is derived from one climate model or from one instrumental data source. Here we find that reconstructions using covariance estimated from combinations of multiple climate models produce less error than reconstructions that use just one climate model. Key Points: Ensembles drawn from single climate models and combinations of multiple models are used to reconstruct surface air temperature variabilityReconstructions from multi‐model ensembles produce lower error than reconstructions from single‐model ensemblesReconstructions from multi‐model ensembles show the largest decreases in error in regions with few observations such as high‐latitude oceans [ABSTRACT FROM AUTHOR]

Paleoclimate reconstruction relies on estimates of spatiotemporal relationships among climate quantities to interpolate between proxy data. This work quantifies how structural uncertainties in those relationships translate to uncertainties in reconstructions of past climate. We develop and apply a data assimilation uncertainty quantification approach to paleoclimate networks and observational uncertainties representative of data for the last millennium. We find that structural uncertainties arising from uncertain spatial covariance relationships typically contribute 10% of the total uncertainty in reconstructed temperature variability at small (∼200 km), continental, and hemispheric length scales, with larger errors (50% or larger) in regions where long‐range climate covariances are least certain. These structural uncertainties contribute far more to errors in uncertainty quantification, sometimes by a factor of 5 or higher. Accounting for and reducing uncertainties in climate model dynamics and resulting covariance relationships will improve paleoclimate reconstruction accuracy. Plain Language Summary: Reconstructing past climate conditions is important to characterize climate variability and extremes, with implications for understanding future variability and change. An important source of information is data from proxy sources like tree rings, ice cores, ocean sediment cores, and other sources, but those data are not available everywhere on the globe. Extracting climate information from these data requires information about how climate properties covary in space and time, and that information is inevitably flawed because we lack a complete picture of how the climate behaves. This work seeks to address how much those flaws matter, how they propagate to reconstructions of past climate, and how we can reduce them. We use different physically plausible representations of climate physics from three different coupled climate models in a series of reconstruction experiments and suggest that uncertain physics typically accounts for 10% of reconstruction errors in the particular case of reconstructing surface air temperature over the Last Millennium, with regionally higher values (50% or larger). We can reduce these errors by improving climate models and by improving how we represent uncertain physics in climate reconstructions. Key Points: Perfect model experiments show how uncertain climate covariance relationships lead to uncertainties in paleoclimate reconstructionsThese uncertainties typically account for 10% of total error in reconstructed Last Millennium air temperature, with large regional variabilityImproving climate model physics and accounting for its uncertainty will improve reconstruction accuracy [ABSTRACT FROM AUTHOR]

Attribution and prediction of global and regional warming requires a better understanding of the magnitude and spatial characteristics of internal global mean surface air temperature (GMST) variability. We examine interdecadal GMST variability in Coupled Modeling Intercomparison Projects, Phases 3, 5, and 6 (CMIP3, CMIP5, and CMIP6) preindustrial control (piControl), last millennium, and historical simulations and in observational data. We find that several CMIP6 simulations show more GMST interdecadal variability than the previous generations of model simulations. Nonetheless, we find that 100‐year trends in CMIP6 piControl simulations never exceed the maximum observed warming trend. Furthermore, interdecadal GMST variability in the unforced piControl simulations is associated with regional variability in the high latitudes and the east Pacific, whereas interdecadal GMST variability in instrumental data and in historical simulations with external forcing is more globally coherent and is associated with variability in tropical deep convective regions. Plain Language Summary: Ongoing and future global and regional warming will progress as a combination of internal climate variability and forced climate change. Understanding the magnitude and spatial patterns associated with internal climate variability is an important aspect of being able to predict when, where, and how climate change will be felt around the globe. Here, we show that the latest climate model simulations, which will be used in the Intergovernmental Panel on Climate Change (IPCC) Assessment Report 6 (AR6), simulate a large range in magnitudes of internal global mean temperature variability. Although there are large unforced global temperature trends in some models, we find that even the most variable models never generate unforced global temperature trends equal to the recently observed global warming trends forced by greenhouse gas emissions. We examine the regions associated with internal climate variability and forced climate change in climate model simulations and find that only forced simulations show a pattern of warming consistent with instrumental data. Key Points: Several CMIP6 control simulations show more interdecadal global mean temperature variability than previous model generationsEven the most variable CMIP6 models never show century‐length global mean temperature trends that exceed observed warming trendsUnlike in control simulations, observed global mean temperature variability is coherent with variability in tropical convective regions [ABSTRACT FROM AUTHOR]