Search for: All records

Abstract Historical observations of Earth’s climate underpin our knowledge and predictions of climate variability and change. However, the observations are incomplete and uncertain, and existing datasets based on these observations typically do not assimilate observations simultaneously across different components of the climate system, yielding inconsistencies that limit understanding of coupled climate dynamics. Here, we use coupled data assimilation, which synthesizes observational and dynamical constraints across all climate fields simultaneously, to reconstruct globally resolved sea surface temperature (SST), near-surface air temperature (T), sea level pressure (SLP), and sea ice concentration (SIC), over 1850–2023. We use a Kalman filter and forecasts from an efficient emulator, the linear inverse model (LIM), to assimilate observations of SST, landT, marine SLP, and satellite-era SIC. We account for model error by training LIMs on eight CMIP6 models, and we use the LIMs to generate eight independent reanalyses with 200 ensemble members, yielding 1600 total members. Key findings in the tropics include post-1980 trends in the Walker circulation that are consistent with past variability, whereas the tropical SST contrast (the difference between warmer and colder SSTs) shows a distinct strengthening since 1975. El Niño–Southern Oscillation (ENSO) amplitude exhibits substantial low-frequency variability and a local maximum in variance over 1875–1910. In polar regions, we find a muted cooling trend in the Southern Ocean post-1980 and substantial uncertainty. Changes in Antarctic sea ice are relatively small between 1850 and 2000, while Arctic sea ice declines by 0.5 ± 0.1 (1σ) million km²during the 1920s. Significance StatementThe key advance in our reconstruction is that the ocean, atmosphere, and sea ice are dynamically consistent with each other and with observations across all components, thus forming a true climate reanalysis. Existing climate datasets are typically derived separately for each component (e.g., atmosphere, ocean, and sea ice), leading to spurious trends and inconsistencies in coupled climate variability. We use coupled data assimilation to unify observations and coupled dynamics across components. We combine forecasts from climate models with observations from ocean vessels and weather stations to produce monthly state estimates spanning 1850–2023 and a novel quantification of globally resolved uncertainty. This reconstruction provides insights into historical variability and trends while motivating future efforts to reduce uncertainties in the climate record. 

Abstract Models from the Coupled Model Intercomparison Project phase 6 (CMIP6) typically struggle to reproduce observed Antarctic sea ice trends, a bias that is substantially alleviated when constraining winds. We use wind‐nudged simulations from two CMIP models to investigate the influence of clouds on sea ice area (SIA). We find that nudging model winds in coupled simulations toward reanalysis, in addition to improving SIA variability, is crucial to reproduce realistic anomalies in cloud radiative effect (CRE) and cloud cover. Biases in the variability of cloud properties at sea ice edge—characterized by CRE anomalies—help explain the remaining discrepancies between simulated and observed SIA; a bias of 1  in the CRE anomaly corresponds to a negative bias of 0.43  in SIA anomaly. Finally, we find that most CMIP6 models show positive trends in CRE anomaly biases, which should contribute to enhanced SIA decline, a long‐standing bias in CMIP models. 

Abstract The variability of Arctic sea ice extent (SIE) on interannual and multidecadal time scales is examined in 29 models with historical forcing participating in phase 6 of the Coupled Model Intercomparison Project (CMIP6) and in twentieth-century sea ice reconstructions. Results show that during the historical period with low external forcing (1850–1919), CMIP6 models display relatively good agreement in their representation of interannual sea ice variability (IVSIE) but exhibit pronounced intermodel spread in multidecadal sea ice variability (MVSIE), which is overestimated with respect to sea ice reconstructions and is dominated by model uncertainty in sea ice simulation in the subpolar North Atlantic. We find that this is associated with differences in models’ sensitivity to Northern Hemispheric sea surface temperatures (SSTs). Additionally, we show that while CMIP6 models are generally capable of simulating multidecadal changes in Arctic sea ice from the mid-twentieth century to present day, they tend to underestimate the observed sea ice decline during the early twentieth-century warming (ETCW; 1915–45). These results suggest the need for an improved characterization of the sea ice response to multidecadal climate variability in order to address the sources of model bias and reduce the uncertainty in future projections arising from intermodel spread. Significance StatementThe credibility of Arctic sea ice predictions depends on whether climate models are capable of reproducing changes in the past climate, including patterns of sea ice variability which can mask or amplify the response to global warming. This study aims to better understand how latest-generation global climate models simulate interannual and multidecadal variability of Arctic sea ice relative to available observations. We find that models differ in their representation of multidecadal sea ice variability, which is overall larger than in observations. Additionally, models underestimate the sea ice decline during the period of observed warming between 1915 and 1945. Our results suggest that, to achieve better predictions of Arctic sea ice, the realism of low-frequency sea ice variability in models should be improved. 

Abstract We evaluate Linear Inverse Models (LIMs) trained on last millennium model data to predict Arctic sea‐ice concentration, thickness, and other atmospheric and oceanic variables on monthly timescales. We find that more than 500 years of training data and 100 years of validation data are needed to reliably estimate LIM forecast skill. The best LIM has skill up to 8 months lead time and outperforms an autoregressive model of order one (AR1) forecast at all locations, with particularly large outperformance near the ice edge. However, for out‐of‐sample validation tests using data from various different model simulations and reanalysis products, they underperform an AR1 model due to differences in the location of the sea‐ice edge from the training data. We present a metric for predicting LIM forecast skill, based on the spatial correlation of the variance in the training and validation data sets. 

Abstract Here we investigate the role of the atmospheric circulation in the Atlantic Meridional Overturning Circulation (AMOC) by comparing a fully‐coupled large ensemble, a forced‐ocean simulation, and new experiments using a fully‐coupled global climate model where winds above the boundary layer are nudged toward reanalysis. When winds are nudged north of 45°N, agreement with RAPID array observations of AMOC at 26.5°N improves across several metrics. The phasing of interannual variability is well‐captured due to the response of the local Ekman component in both wind‐nudging and forced‐ocean simulations, however the variance remains underestimated. The mean AMOC strength is substantially reduced relative to the fully‐coupled model large ensemble, which is biased high, due to the impact of winds on surface buoyancy fluxes over the subpolar gyre. Nudging winds toward observations also reduces the 1979–2016 trend in AMOC, suggesting that improvement in the representation of the high‐latitude atmosphere is important for projecting long‐term AMOC changes. 

Biomass burning can affect climate via the emission of aerosols and their subsequent impact on radiation, cloud microphysics, and surface and atmospheric albedo. Biomass burning emissions (BBEs) over the boreal region have strongly increased during the last decade and are expected to continue increasing as the climate warms. Climate models simulate aerosol processes, yet historical and future Coupled Model Intercomparison Project (CMIP) simulations have no active fire component, and BBEs are prescribed as external forcings. Here, we show that CMIP6 used future boreal BBEs scenarios with unrealistic near-zero trends that have a large impact on climate trends. By running sensitivity experiments with ramped up boreal emissions based on observed trends, we find that increasing boreal BBEs reduces global warming by 12% and Arctic warming by 38%, reducing the loss of sea ice. Tropical precipitation shifts southward as a result of the hemispheric difference in boreal aerosol forcing and subsequent temperature response. These changes stem from the impact of aerosols on clouds, increasing cloud droplet number concentration, cloud optical depth, and low cloud cover, ultimately reducing surface shortwave flux over northern latitudes. Our results highlight the importance of realistic boreal BBEs in climate model simulations and the need for improved understanding of boreal emission trends and aerosol–climate interactions. 

The observed rate of global warming since the 1970s has been proposed as a strong constraint on equilibrium climate sensitivity (ECS) and transient climate response (TCR)—key metrics of the global climate response to greenhouse-gas forcing. Using CMIP5/6 models, we show that the inter-model relationship between warming and these climate sensitivity metrics (the basis for the constraint) arises from a similarity in transient and equilibrium warming patterns within the models, producing an effective climate sensitivity (EffCS) governing recent warming that is comparable to the value of ECS governing long-term warming under CO ${}_2$forcing. However, CMIP5/6 historical simulations do not reproduce observed warming patterns. When driven by observed patterns, even high ECS models produce low EffCS values consistent with the observed global warming rate. The inability of CMIP5/6 models to reproduce observed warming patterns thus results in a bias in the modeled relationship between recent global warming and climate sensitivity. Correcting for this bias means that observed warming is consistent with wide ranges of ECS and TCR extending to higher values than previously recognized. These findings are corroborated by energy balance model simulations and coupled model (CESM1-CAM5) simulations that better replicate observed patterns via tropospheric wind nudging or Antarctic meltwater fluxes. Because CMIP5/6 models fail to simulate observed warming patterns, proposed warming-based constraints on ECS, TCR, and projected global warming are biased low. The results reinforce recent findings that the unique pattern of observed warming has slowed global-mean warming over recent decades and that how the pattern will evolve in the future represents a major source of uncertainty in climate projections.