Trends in Arctic sea ice extent from CMIP5, CMIP3 and observations

Both the Arctic and Antarctic sea ice extents (SIEs) from 44 coupled models in the Coupled Model Intercomparison Project Phase 6 (CMIP6) are evaluated by comparing them with observations and CMIP5 results. The CMIP6 multimodel mean can adequately reproduce the seasonal cycles of both the Arctic and Antarctic SIE. The observed Arctic September SIE declining trend (−0.82 ± 0.18 million km2 per decade) between 1979 and 2014 is slightly underestimated in CMIP6 models (−0.70 ± 0.06 million km2 per decade). The observed weak but significant upward trend of the Antarctic SIE is not captured, which was an issue already in the CMIP5 phase. Compared with CMIP5 models, CMIP6 models have lower intermodel spreads in SIE mean values and trends, although their SIE biases are relatively larger. The CMIP6 models did not reproduce the new summer tendencies after 2000, including the faster decline of Arctic SIE and the larger interannual variability in Antarctic SIE.

Quantitative estimates of the relationship between interannual variations in the extent of Antarctic and Arctic sea ice and changes in the surface air temperature in the Northern and Southern hemispheres are obtained using satellite, ground-based, and reanalysis data for the past four decades (1980–2019). It is shown that the previously noted general increase in the extent of Antarctic sea ice observed until recent years from satellite data (available only since the late 1970s) over the background global warming and a rapid decrease in the extent of Arctic sea ice is associated with a regional decrease in the surface temperature at Antarctic latitudes from the end of the 1970s. This is a result of regional manifestation of natural climate variations with periods of up to several decades against the background of global secular warming with a relatively weak temperature trend over the ocean in the Southern Hemisphere. Since 2016, a sharp decrease in the extent of Antarctic sea ice in the Southern Ocean has been observed. The results of the correlation and cross-wavelet analysis indicate significant coherence and negative correlation with the surface temperature of the extent of sea ice in recent decades, not only in the Arctic, but also in the Antarctic.

Abstract. In response to increasing greenhouse gases, the subtropical edges of Earth's Hadley circulation shift poleward in global climate models. Recent studies have found that reanalysis trends in the Hadley cell edge over the past 30–40 years are within the range of trends simulated by Coupled Model Intercomparison Project Phase 5 (CMIP5) models and have documented seasonal and hemispheric asymmetries in these trends. In this study, we evaluate whether these conclusions hold for the newest generation of models (CMIP6). Overall, we find similar characteristics of Hadley cell expansion in CMIP5 and CMIP6 models. In both CMIP5 and CMIP6 models, the poleward shift of the Hadley cell edge in response to increasing greenhouse gases is 2–3 times larger in the Southern Hemisphere (SH), except during September–November. The trends from CMIP5 and CMIP6 models agree well with reanalyses, although prescribing observed coupled atmosphere–ocean variability allows the models to better capture reanalysis trends in the Northern Hemisphere (NH). We find two notable differences between CMIP5 and CMIP6 models. First, while both CMIP5 and CMIP6 models contract the NH summertime Hadley circulation equatorward (particularly over the Pacific sector), this contraction is larger in CMIP6 models due to their higher average climate sensitivity. Second, in recent decades, the poleward shift of the NH annual-mean Hadley cell edge is slightly larger in CMIP6 models. Increasing greenhouse gases drive similar trends in CMIP5 and CMIP6 models, so the larger recent NH trends in CMIP6 models point to the role of other forcings, such as aerosols.

Abstract. With the ongoing decline in Arctic sea ice extent, the accurate simulation of Arctic sea ice in coupled models remains an important problem in climate modelling. In this study, the substantial Coupled Model Intercomparison Project Phase 6 (CMIP6) model spread in Arctic sea ice extent and volume is investigated using a novel, process-based approach. An observational dataset derived from the Arctic ice mass balance buoy (IMB) network is used to evaluate fluxes of melt, growth and conduction produced by a subset of CMIP6 models, to better understand the model processes that underlie the large-scale sea ice states. Due to the sparse nature of the IMB observations, the evaluation is performed by comparing distributions of modelled and observed fluxes in the densely sampled regions of the North Pole and Beaufort Sea. We find that all fluxes are routinely biased high in magnitude with respect to the IMB measurements by nearly all models, with too much melt in summer and too much conduction and growth in winter, even as a function of ice thickness. We also show that fluxes vary in ways which are physically consistent with the thermodynamic parameterisations used and that these effects likely modulate the large-scale relationship between ice thickness and ice growth and melt in the CMIP6 models.

The double‐intertropical convergence zone (ITCZ) bias is one of the most outstanding errors in all previous generations of climate models. Here, the annual double‐ITCZ bias and the associated precipitation bias in the latest climate models for Coupled Model Intercomparison Project (CMIP) Phase 6 (CMIP6) are examined in comparison to their previous generations (CMIP Phase 3 [CMIP3] and CMIP Phase 5 [CMIP5]). All three generations of CMIP models share similar systematic annual multi‐model ensemble mean precipitation errors in the tropics. The notorious double‐ITCZ bias and its big inter‐model spread persist in CMIP3, CMIP5, and CMIP6 models. Based on several tropical precipitation bias indices, the double‐ITCZ bias is slightly reduced from CMIP3 or CMIP5 to CMIP6. In addition, the annual equatorial Pacific cold tongue persists in all three generations of CMIP models, but its inter‐model spread is reduced from CMIP3 to CMIP5 and from CMIP5 to CMIP6.

Radley Horton,1,a Daniel Bader,1,a Yochanan Kushnir,2 Christopher Little,3 Reginald Blake,4 and Cynthia Rosenzweig5 1Columbia University Center for Climate Systems Research, New York, NY. 2Ocean and Climate Physics Department, Lamont-Doherty Earth Observatory, Columbia University, Palisades, NY. 3Atmospheric and Environmental Research, Lexington, MA. 4Physics Department, New York City College of Technology, CUNY, Brooklyn, NY. 5Climate Impacts Group, NASA Goddard Institute for Space Studies; Center for Climate Systems Research, Columbia University Earth Institute, New York, NY

Abstract. Simulation of the carbon cycle in climate models is important due to its impact on climate change, but many weaknesses in its reproduction were found in previous models. Improvements in the representation of the land carbon cycle in Earth system models (ESMs) participating in the Coupled Model Intercomparison Project Phase 6 (CMIP6) include the interactive treatment of both the carbon and nitrogen cycles, improved photosynthesis, and soil hydrology. To assess the impact of these model developments on aspects of the global carbon cycle, the Earth System Model Evaluation Tool (ESMValTool) is expanded to compare CO2-concentration- and CO2-emission-driven historical simulations from CMIP5 and CMIP6 to observational data sets. A particular focus is on the differences in models with and without an interactive terrestrial nitrogen cycle. Overestimations of photosynthesis (gross primary productivity (GPP)) in CMIP5 were largely resolved in CMIP6 for participating models with an interactive nitrogen cycle but remaining for models without one. This points to the importance of including nutrient limitation. Simulating the leaf area index (LAI) remains challenging, with a large model spread in both CMIP5 and CMIP6. In ESMs, the global mean land carbon uptake (net biome productivity (NBP)) is well reproduced in the CMIP5 and CMIP6 multi-model means. However, this is the result of an underestimation of NBP in the Northern Hemisphere, which is compensated by an overestimation in the Southern Hemisphere and the tropics. Carbon stocks remain a large uncertainty in the models. While vegetation carbon content is slightly better represented in CMIP6, the inter-model range of soil carbon content remains the same between CMIP5 and CMIP6. Overall, a slight improvement in the simulation of land carbon cycle parameters is found in CMIP6 compared to CMIP5, but with many biases remaining, further improvements of models in particular for LAI and NBP is required. Models from modeling groups participating in both CMIP phases generally perform similarly or better in their CMIP6 compared to their CMIP5 models. This improvement is not as significant in the multi-model means due to more new models in CMIP6, especially those using older versions of the Community Land Model (CLM). Emission-driven simulations perform just as well as the concentration-driven models, despite the added process realism. Due to this, we recommend that ESMs in future Coupled Model Intercomparison Project (CMIP) phases perform emission-driven simulations as the standard so that climate–carbon cycle feedbacks are fully active. The inclusion of the nitrogen limitation led to a large improvement in photosynthesis compared to models not including this process, suggesting the need to view the nitrogen cycle as a necessary part of all future carbon cycle models. Possible benefits when including further limiting nutrients such as phosphorus should also be considered.

<p>Over the past decades, Arctic sea ice has declined in thickness and extent and is shifting toward a seasonal ice regime. These rapid changes have widespread implications for ecological and human activities as well as the global climate, and accurate predictions could benefit a wide range of stakeholders, from local residents to governmental policy makers. However, many aspects of the polar transient climate response remain poorly understood, particularly in regard to the response of Arctic sea ice to increasing atmospheric CO<sub>2</sub> concentration and warming temperatures. The Coupled Model Intercomparison Project Phase 6 (CMIP6) provides a useful framework for understanding this response, and the participating climate model simulations are a powerful tool for advancing our understanding of present and future changes in the Arctic climate system.</p><p>Here we explore the current and future states of Arctic sea ice in the Community Earth System Model version 2 (CESM2), the latest generation of the CESM and NCAR’s contribution to CMIP6. We analyze changes in Arctic sea ice cover in two CESM2 configurations with differing atmospheric components: the “low-top” configuration with limited chemistry (CESM2-CAM) and the “high-top” configuration with interactive chemistry (CESM2-WACCM). We find that the two experiments show large differences in their simulation of Arctic sea ice over the historical period. The CESM2-CAM winter ice thickness distribution is skewed thin, with an insufficient amount of ice thicker than 3 m. This leads to a lower summer ice extent compared to the CESM2-WACCM and observations. In both experiments, the timing of first ice-free conditions is insensitive to the choice of future emissions scenario (known as the shared socioeconomic pathways, or SSPs, in CMIP6), an alarming result that points to the current vulnerable state of Arctic sea ice. However, if global warming stays below 1.5°C, the probability of an ice-free summer remains low, consistent with other recent studies. By the end of the 21<sup>st</sup> century, both experiments exhibit an accelerated decline in winter ice extent under the high emissions scenario (SSP5-8.5), leading to ice-free conditions for up to 8 months and an open-water period of 220 days or more depending on the region. Initial results show that the CESM2 simulates less ocean heat loss during the fall months compared to its previous version, delaying the formation of sea ice and leading to lower winter ice extent. Given that the CESM2 reaches a higher atmospheric CO<sub>2</sub> concentration and thus warmer global and Arctic temperatures by 2100, these results suggest the presence of emerging processes associated with a state of the Arctic climate that has never been sampled before.</p>

This paper is focused on the seasonality change of Arctic sea ice extent (SIE) from 1979 to 2100 using newly available simulations from the Coupled Model Intercomparison Project Phase 5 (CMIP5). A new approach to compare the simulation metric of Arctic SIE between observation and 31 CMIP5 models was established. The approach is based on four factors including the climatological average, linear trend of SIE, span of melting season and annual range of SIE. It is more objective and can be popularized to other comparison of models. Six good models (GFDL-CM3, CESM1-BGC, MPI-ESM-LR, ACCESS-1.0, HadGEM2-CC, and HadGEM2-AO in turn) are found which meet the criterion closely based on above approach. Based on ensemble mean of the six models, we found that the Arctic sea ice will continue declining in each season and firstly drop below 1 million km2 (defined as the ice-free state) in September 2065 under RCP4.5 scenario and in September 2053 under RCP8.5 scenario. We also study the seasonal cycle of the Arctic SIE and find out the duration of Arctic summer (melting season) will increase by about 100 days under RCP4.5 scenario and about 200 days under RCP8.5 scenario relative to current circumstance by the end of the 21st century. Asymmetry of the Arctic SIE seasonal cycle with later freezing in fall and early melting in spring, would be more apparent in the future when the Arctic climate approaches to “tipping point”, or when the ice-free Arctic Ocean appears. Annual range of SIE (seasonal melting ice extent) will increase almost linearly in the near future 30–40 years before the Arctic appears ice-free ocean, indicating the more ice melting in summer, the more ice freezing in winter, which may cause more extreme weather events in both winter and summer in the future years.

The Arctic has experienced rapid sea ice loss and a substantial surface albedo decline, altering its radiation budget. CMIP6 (Coupled Model Intercomparison Project Phase 6) models capture these trends but show considerable inter‐model spread in the magnitude, distribution, and seasonality of Arctic surface albedo. Over land, spread in AMIP (Atmospheric Model Intercomparison Project) and CMIP6 is associated with snow cover variations, while over the ocean, where sea ice dominates, the sources are less clear. We compare CMIP6 simulations with observations from the Clouds and the Earth's Radiant Energy System (CERES) and develop a decomposition method to quantify contributions from sea ice albedo, concentration, and extent. Over the Arctic Ocean, all three factors contribute to inter‐model spread in CMIP6. In AMIP, despite prescribed sea ice concentrations, we were surprised to find an inter‐model spread in Arctic Ocean albedo similar to that in CMIP6, driven solely by sea ice albedo. Applying the decomposition to future projections shows that the largest decline occurs in the Central Arctic, driven primarily by reductions in sea ice extent. After 2045, sea ice extent emerges as the dominant driver, highlighting ice edge retreat as key to future albedo decline. These findings pinpoint key sources of inter‐model spread in Arctic surface albedo and offer insights into quantifying shortwave (SW) radiative effect associated with sea ice responses in future projections.

Using state‐of‐the‐art models from the Coupled Model Intercomparison Project phase 5 (CMIP5), this study found the upper Arctic Ocean likely to freshen considerably in the future. Arctic Ocean average sea surface salinity is projected to decrease by 1.5 ± 1.1 psu, and the liquid freshwater column is projected to increase by 5.4 ± 3.8 m by the end of the 21st century under the Representative Concentration Pathway 8.5 (RCP8.5) scenario. Most freshening would occur in the Arctic Ocean basins, that is, the Canada, Makarov, and Amundsen basins. Anomalies in freshwater flux from sea ice melt, Bering Strait inflow, net precipitation (P‐E), river runoff, and freshwater through the Barents Sea Opening (BSO) would contribute to Arctic Ocean freshening. CMIP5 historical and RCP8.5 experiments showed that the respective projected contributions from BSO freshwater flux, river runoff, P‐E, and Bering Strait inflow are about 6.4, 5.0, 2.7, and 2.2 times the contribution from sea ice melt averaged throughout the 21st century. Contributions from sea ice melt and Bering Strait inflow would increase and then decrease gradually, while those from BSO freshwater flux, river runoff, and P‐E would increase continuously. The CMIP5 models are able to simulate the Arctic Ocean freshwater system more accurately than CMIP3 models. However, the simulated rate of increase of freshwater content (296 ± 232 km3/yr) is weaker than estimated (600 ± 300 km3/yr) based on observations (1992–2012). Moreover, the simulated BSO and Davis Strait freshwater fluxes still exhibit substantial intermodel spread and they differ considerably from observed values.

A comparison assessment of model capabilities in simulating precipitation extremes across China was first implemented by using 30 models from the Coupled Model Intercomparison Project Phase 5 (CMIP5) and using 36 CMIP6 models. The results indicate that the multi‐model median ensembles (MME) of both the CMIP5 and CMIP6 models can reasonably reproduce the climate means for the period from 1986 to 2005, and the biases are lower in most CMIP6 models compared to the CMIP5 models, especially over southern China. To provide further comparisons, 14 CMIP6 models are selected and compared with their predecessors in CMIP5. The results show that the CMIP6 models generally exhibit superior skill in simulating the extreme precipitation indices over China. The model spreads for most of the extreme indices in the CMIP6 version are also smaller. Additionally, the MMEs of the two CMIPs outperform individual models. However, some CMIP6 models also exhibit weaker skill levels in simulating some particular indices compared with those in CMIP5, which merits further investigation. The results from seven reanalyses further show large uncertainties for these indices; therefore, care should be taken in comparison with reanalyses. For future changes in precipitation extremes, total wet day precipitation (PRCPTOT), maximum 5‐day precipitation (RX5day) and very heavy precipitation days (R20mm) are projected to clearly increase across China over the coming century under the shared socioeconomic pathway (SSP) 2‐4.5 and SSP5‐8.5 scenarios. However, the dry condition index of CDD exhibits a decreasing tendency in the future, which implies that the dry conditions induced by precipitation anomalies will be mitigated. However, large uncertainties are still observed for future changes, which are primarily sourced from inter‐model and scenario variabilities, especially for the projected changes at the end of the 21st century.

Comparison of Arctic and Antarctic sea ice spatial–temporal changes during 1979–2018

We summarize the latest results on the rapid changes that are occurring to Arctic sea ice thickness and extent, the reasons for them, and the methods being used to monitor the changing ice thickness. Arctic sea ice extent had been shrinking at a relatively modest rate of 3-4% per decade (annually averaged) but after 1996 this speeded up to 10% per decade and in summer 2007 there was a massive collapse of ice extent to a new record minimum of only 4.1 million km(2). Thickness has been falling at a more rapid rate (43% in the 25 years from the early 1970s to late 1990s) with a specially rapid loss of mass from pressure ridges. The summer 2007 event may have arisen from an interaction between the long-term retreat and more rapid thinning rates. We review thickness monitoring techniques that show the greatest promise on different spatial and temporal scales, and for different purposes. We show results from some recent work from submarines, and speculate that the trends towards retreat and thinning will inevitably lead to an eventual loss of all ice in summer, which can be described as a 'tipping point' in that the former situation, of an Arctic covered with mainly multi-year ice, cannot be retrieved.

Abstract. The uptake of anthropogenic carbon (Cant) by the ocean leads to ocean acidification, causing the reduction of pH and the saturation states of aragonite (Ωarag) and calcite (Ωcalc). The Arctic Ocean is particularly vulnerable to ocean acidification due to its naturally low pH and saturation states and due to ongoing freshening and the concurrent reduction in total alkalinity in this region. Here, we analyse ocean acidification in the Arctic Ocean over the 21st century across 14 Earth system models (ESMs) from the latest Coupled Model Intercomparison Project Phase 6 (CMIP6). Compared to the previous model generation (CMIP5), models generally better simulate maximum sea surface densities in the Arctic Ocean and consequently the transport of Cant into the Arctic Ocean interior, with simulated historical increases in Cant in improved agreement with observational products. Moreover, in CMIP6 the inter-model uncertainty of projected changes over the 21st century in Arctic Ocean Ωarag and Ωcalc averaged over the upper 1000 m is reduced by 44–64 %. The strong reduction in projection uncertainties of Ωarag and Ωcalc can be attributed to compensation between Cant uptake and total alkalinity reduction in the latest models. Specifically, ESMs with a large increase in Arctic Ocean Cant over the 21st century tend to simulate a relatively weak concurrent freshening and alkalinity reduction, while ESMs with a small increase in Cant simulate a relatively strong freshening and concurrent total alkalinity reduction. Although both mechanisms contribute to Arctic Ocean acidification over the 21st century, the increase in Cant remains the dominant driver. Even under the low-emissions Shared Socioeconomic Pathway 1-2.6 (SSP1-2.6), basin-wide averaged Ωarag undersaturation in the upper 1000 m occurs before the end of the century. While under the high-emissions pathway SSP5-8.5, the Arctic Ocean mesopelagic is projected to even become undersaturated with respect to calcite. An emergent constraint identified in CMIP5 which relates present-day maximum sea surface densities in the Arctic Ocean to the projected end-of-century Arctic Ocean Cant inventory is found to generally hold in CMIP6. However, a coincident constraint on Arctic declines in Ωarag and Ωcalc is not apparent in the new generation of models. This is due to both the reduction in Ωarag and Ωcalc projection uncertainty and the weaker direct relationship between projected changes in Arctic Ocean Cant and changes in Ωarag and Ωcalc.