Cloud radiative effect dominates variabilities of surface energy budget in the dark Arctic

Cheng Tao; Minghua Zhang; Shaocheng Xie

doi:10.1038/s41598-025-86322-2

. 2025 Jan 23;15:2976. doi: 10.1038/s41598-025-86322-2

Cloud radiative effect dominates variabilities of surface energy budget in the dark Arctic

Cheng Tao ^1,^✉, Minghua Zhang ^2,^✉, Shaocheng Xie ¹

PMCID: PMC11757706 PMID: 39849030

Abstract

Climate models simulate a wide range of temperatures in the Arctic. Here we investigate one of the main drivers of changes in surface temperature: the net surface heat flux in the models. We show that in the winter months of the dark Arctic, there is a more than two-fold difference in the net surface heat fluxes among the models, and this difference is dominated by the downward infrared radiation from clouds. Owing to the small amount of water vapor in the winter Arctic, infrared radiation from clouds transmits more easily to the surface in the Arctic than at other latitudes, resulting in large cloud radiative effect at the surface. The dominant role of the cloud effect is also found in the transient variability of the net surface heat flux. Results demonstrate that accurate simulation of clouds is crucial for determining the net surface heat flux, which in turn affects surface temperature and sea ice properties in the Arctic.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-86322-2.

Subject terms: Climate sciences, Environmental sciences

Introduction

It is widely known that there are considerable uncertainties in the simulated temperature changes of the Arctic in climate models^1–3. Considering that sea ice essentially melts at a fixed temperature, this model discrepancy in Arctic temperature directly contributes to the large spread of simulated sea ice decline in the last three decades and in the future^4,5. It also impacts the timing and magnitude of the Arctic amplification of global warming^1,6, as well as the associated effects on climate and weather in the middle latitudes^7–9. Reducing these uncertainties is a necessary step to gain confidence in the projected future changes in sea ice of the Arctic.

Several factors contribute to the variabilities of sea ice temperature. These include the net surface heat flux, the vertical distribution of temperature, and the growth and melting at the bottom of sea ice¹⁰. Here we focus on the net surface heat flux in the winter months of the Arctic, defined as December to February months (DJF) over areas north of 80°N. This is the season when sea ice accumulates. The surface loses energy due to the emission of infrared radiation to the atmosphere and the space. The surface also exchanges energy with the atmosphere in the form of sensible and latent heat fluxes. The absence of insolation in the winter months allows us to probe into the role of other processes beyond the impact of the absorption of solar radiation.

The impact of cloud radiative effect

Among the 41 models that simulated the present climate in Phase 6 of the Coupled Model Intercomparison Project (CMIP6¹¹, “Methods and Data” Section), the net surface heat flux ( Inline graphic ) ranges from 15 to 40 W/m² (Fig. 1a). There is a discrepancy of about 100% of the mean of 25 W/m². If the net surface heat flux were the only factor affecting the temperature of a slab of 3 m of sea ice that is typical of the winter Arctic^4,12, it would cause about 40 K difference in sea ice temperature over a three-month period. The net loss of heat is primarily due to the net surface infrared radiation ( Inline graphic ) (Fig. 1b). Surface sensible heat flux slightly warms the surface (downward) in most of the models, indicating that air temperature in the lower troposphere is generally warmer than that of the sea ice. Latent heat flux at the surface is negligible because of the low temperature and thus low saturation water vapor pressure. The actual surface temperature in the models varies widely from 236 to 249 K, with a spread of 13 K (Fig. 1c). The mean temperature of 245 K is about 4 K colder than that in ERA5 reanalysis¹³ (the latter of which has been shown to have a warm bias¹⁴). The model spread of 13 K is larger than the averaged projected warming at the end of the 21st century^1,5. Because the Arctic sea ice temperature is not in equilibrium state in the winter, the heat flux is not expected to directly correlate with the sea ice temperature, but it should be one of the main drivers of the temporal change of temperature¹⁰.

Fig. 1 — Simulated surface fields by 41 CMIP6 models and in ERA5. (a) Net surface heat flux (, W/m²). (b) Net surface longwave radiation (), net surface shortwave radiation (), surface latent () and sensible () heat fluxes (W/m²). (c) Surface temperature (K). Values are averaged for December, January, and February of 2005–2014 from the AMIP simulations or ERA5 reanalysis. We consider upward fluxes as positive unless otherwise specified.

The Inline graphic is significantly correlated with in the models (r = 0.72, p < 0.01, Fig. 2a). The can be dissected in two ways. One is the decomposition into the difference between the upward () and downward () fluxes; the other is into the difference between clear-sky net flux () and the cloud radiative effect (CRE)¹⁵ (“Methods and Data” Section). With the first decomposition, we find that Inline graphic is significantly correlated with (r = − 0.43, p < 0.01) but not with (r = 0.12, p = 0.45). With the second decomposition, is significantly correlated with CRE (r = − 0.62, p < 0.01, Fig. 2b) but not with (r = 0.08, p = 0.60). The dependence of the net heat flux on the sensible heat flux is insignificant (r = 0.18, p = 0.27).

Surface temperature is not statistically correlated with Inline graphic (r = 0.16, p = 0.32). This is counter intuitive, warmer surface temperature does not translate to more heat loss from longwave emission at the surface. This is because of the dominance of the impact of clouds. Surface temperature is only correlated with (r = 0.93, p < 0.01, Fig. S1), through which it impacts the net surface heat flux and thus contributes to the differences among the models. When we normalize Inline graphic by using the same surface temperature (“Methods and Data” Section), we find that the direct correlation of with CRE reaches 0.77 at p < 0.01 (Fig. 2c).

The dominance of CRE on the downward longwave radiation is a special feature of clouds in regions where water vapor concentration is low. The infrared radiation emitted downward from clouds is partially absorbed by water vapor and CO₂ in the atmospheric layer between the clouds and surface. In the dark Arctic, the average column-integrated water vapor in the models is around 2.5 mm, which is about 10% of the global average of about 24 mm. Water vapor has a strong absorption band from vibrational transitions around 6.3 μm and many vibrational-rotational absorption lines across the infrared spectrum. In addition, it absorbs radiation in its continuum in the far infrared around 20 μm. CO₂ has a strong vibrational absorption band at around 15 μm. We compared the transmittance of infrared radiation through a layer of 2 km atmosphere with column water vapor of 2.5 mm to that of a layer with 24 mm from the US Standard Atmosphere¹⁶, assuming CO₂ concentration of 400 ppmv. We find that the transmittance is significantly larger for the Arctic layer across the whole infrared spectra (Fig. 2d versus e). Specifically, the spectral ranges of the opaque atmosphere around the centers of the 6.3 μm water vapor absorption band and the 15 μm CO₂ absorption band are narrower for the Arctic layer than that for the layer with more water vapor. The transmittance in the window region of about 7 μm to 14 μm and in the far-infrared beyond 17 μm is considerably larger in the Arctic (Fig. 2e). Using the range of 17 μm to 18 μm as an example, we find that for the 8 absorption lines of water vapor in the HITRAN 2020 database¹⁷, the standard atmosphere is about 90% opaque to infrared radiation with near-zero transmittance (red line in Fig. 2f). In contrast, the Arctic atmosphere is considerably more transparent (blue line in Fig. 2f), with the wings of these absorption lines largely open for transmission of infrared radiation. Therefore, a considerable fraction of emitted infrared radiation from Arctic clouds can reach the surface to affect the net energy budget, while it is largely blocked by the atmosphere in the US Standard Atmosphere.

Transient variability

The dominance of the cloud radiative effect on the variability of the surface longwave radiation and the net surface heat flux is corroborated from the analysis of atmospheric data on shorter time scales. Using the hourly ERA5 reanalysis during the three months from November 2019 to January 2020 in the Multidisciplinary drifting Observatory for the Study of Arctic Climate (MOSAiC) field campaign^18,19 (“Methods and Data” Section), we find that Inline graphic is also significantly correlated with (r = 0.86, p < 0.01, Fig. 3a). The variability of is in turn dominated by CRE (r = − 0.94, p < 0.01, Fig. 3b). The direct correlation of with the CRE is also significant (r = − 0.76, p < 0.01, Fig. 3c).

Fig. 3 — (a–c), Same as Fig. 2a–c except that hourly ERA5 data during the winter months (December 2019 to February 2020) of the MOSAiC field campaign are used, and that the net surface heat flux in (c) is not normalized (because there are sufficient samples of data). (d) Cloud amount (color image), downward longwave radiation at the surface (red line), clear-sky downward longwave radiation at the surface (white line), and cloud radiative effect for November 2019 in ERA5 (black line) (W/m²). (e) Same as (d) but from the ARM Best Estimate dataset. Results in (d) and (e) represent the same location following the ship-based moving observational platforms. The clear-sky radiation is calculated by using balloon sounding data and the radiative transfer model RRTMG.

Clouds over sea ice in the dark Arctic fall into one of the two primary categories: shallow stratus in ice and mixed phase^20,21, and deep clouds associated with cyclones^22–25. ERA5 captures both types as seen in a time-pressure cross section of clouds for the month of November 2019 (Fig. 3d). The maxima of downward longwave radiation in ERA5 are all collocated with the presence of these clouds (Fig. 3d, red line), demonstrating the penetration of infrared radiation to the surface. As can be expected, the thermal profiles of temperature and water vapor also affect the surface radiation, as indicated by the clear-sky downward radiation at the surface (Fig. 3d, white line). The large variation of the downward longwave radiation, however, is due to the change of CRE (Fig. 3d, black line). This is consistent with previous studies that showed the association of low-level clouds with longwave radiation^26,27.

During the MOSAiC field campaign, radars and lidars were deployed to measure the vertical profiles of clouds. Collocated radiometers were used to measure the downward longwave radiation at the surface (“Methods and Data” Section). The variabilities of the cloud profiles from the remote sensing instruments have clear imprints on the downward longwave radiation at the surface (Fig. 3e, red line), which mirrors the variation of the CRE (Fig. 3e, black line).

As a comparison, results from the March 2000 Intensive Observational Period (IOP) field campaign of the Atmospheric Radiation Measurement Program (ARM) at the Southern Great Plains (SGP)²⁸ are also examined. This IOP features both low-level clouds and frontal clouds as for the MOSAiC period (Fig. S2). The dependence of downward longwave radiation on the CRE at SGP is much less significant compared with that in MOSAiC.

The close correspondence of the CRE with the net surface heat flux is applicable to other locations where the water vapor concentration is low. This can be seen from the spatial distribution of the direct correlation of the hourly Inline graphic for the winter months with the longwave CRE in ERA5 (Fig. 4a) and in the five selected CMIP6 models (Fig. 4b–f). The negative correlation in the deep Arctic indicates that when CRE is large, there is less loss of heat from the surface. The correlation patterns vary from model to model. This can be explained by the differences of cloud properties and water vapor among other factors in the models.

Fig. 4 — Spatial distributions of the correlation coefficient between the net surface heat flux () and the surface longwave cloud radiative effect (CRE) for (a) ERA5, (b) CESM2, (c) E3SM-1-0, (d) GFDL-CM4, (e) GISS-E2-1-G, and (f) UKESM1-0-LL. Only the correlation coefficient with a 95% confidence (p < 0.05) are shown.

Discussions

Clouds reflect solar radiation to cool the planet while they warm the planet via their greenhouse effects through their impact on the radiative fluxes at the top of the atmosphere. The variation of clouds leads to climate feedbacks that is the primary reason for discrepancies in the sensitivity of climate models^29–31. Here we present a different role of clouds: They cause large discrepancies of the net surface energy budget among the models in the winter polar regions.

The large discrepancy of downward longwave radiation in the Arctic has been previously shown in seven state-of-the-art operational and experimental forecasting systems³² and in other models³³.

In the last two decades for the winter Arctic, opposite trends have been reported in the surface radiation from reanalysis and satellite observations, which are dominated by the downward longwave radiation³⁴. Our study further attributes these differences to cloud radiative effects as a result of the large atmospheric transmittance in the infrared due to the low concentration of water vapor in the Arctic.

Accurate simulation of clouds and their radiative effects are therefore necessary to calculate the net surface heat flux, thus the surface temperature and sea ice properties in the Arctic. Unlike the global cloud feedback problem that involves radiative flux measurements at the top of the atmosphere over the whole globe, surface radiation and collocated clouds in the polar region can be better measured. Observations such as those from MOSAiC can be used to evaluate and improve model performance. Our results call for more attention to clouds in the polar regions to improve the models.

We have not addressed what causes the model differences in clouds, and thus their radiative effects. The deep clouds shown in Fig. 3 are associated with intrusion of synoptic weather systems, while the shallow clouds are primarily associated with boundary layer processes. These can differ greatly in the models. Surface temperature itself in the polar region is a contributor to these differences. By revealing the impact of clouds on the surface energy budget, and ultimately the changes of surface temperature, we expose one target for models to improve. An interesting question is how polar clouds may change in the future and thus feedback to temperature. Because polar clouds change in a warming climate, the present study implies that clouds have the potential to amplify or mitigate future changes in polar temperature.

Methods and data

Datasets

CMIP models.

Monthly outputs of relevant surface energy budget components from 41 CMIP6¹¹ models in their Atmospheric Model Intercomparison Project (AMIP) experiments were analyzed (Table 1). These experiments use prescribed sea surface temperature and sea ice concentrations from observations. Here, we use the first available ensemble member for each model and focus on the winter months of the dark Arctic, i.e., December-January-February of 2005–2014 and north of 80°N.

Table 1.

CMIP6 models analyzed in this study.

Models	Institute	Horizontal grid size (lon × lat)
ACCESS-CM2	Commonwealth Scientific and Industrial Research Organization (CSIRO) and Bureau of Meteorology (BoM), Australia	192 × 144
ACCESS-ESM1-5		192 × 145
BCC-CSM2-MR	Beijing Climate Center, China	320 × 160
CanESM5	Canadian Centre for Climate Modelling and Analysis, Canada	128 × 64
CAS-ESM2-0	Institute of Atmospheric Physics, Chinese Academy of Sciences, China	256 × 128
FGOALS-f3-L		288 × 180
FGOALS-g3		180 × 80
CESM2	National Center for Atmospheric Research, USA	288 × 192
CESM2-FV2		144 × 96
CESM2-WACCM		288 × 192
CESM2-WACCM-FV2		144 × 96
CIESM	Tsinghua University, Beijing, China	288 × 192
CMCC-CM2-SR5	Euro-Mediterranean Center on Climate Change (CMCC), Italy	288 × 192
CNRM-CM6-1	Centre National de Recherches Météorologiques, France	256 × 128
CNRM-CM6-1-HR		720 × 360
CNRM-ESM2-1		256 × 128
E3SM-1-0	U.S. Department of Energy (DOE)	360 × 180
EC-Earth3	European Centre for Medium-Range Weather Forecasts (ECMWF)	512 × 256
EC-Earth3-AerChem		512 × 256
EC-Earth3-CC		512 × 256
EC-Earth3-Veg		512 × 256
EC-Earth3-Veg-LR		320 × 160
GFDL-AM4	NOAA Geophysical Fluid Dynamics Laboratory	288 × 180
GFDL-CM4	NOAA Geophysical Fluid Dynamics Laboratory	288 × 180
GISS-E2-1-G	NASA Goddard Institute for Space Studies, USA	144 × 90
HadGEM3-GC31-LL	Met Office Hadley Centre, Exeter, UK	192 × 144
UKESM1-0-LL	Met Office Hadley Centre, Exeter, UK	192 × 144
IITM-ESM	Indian Institute of Tropical Meteorology, India	192 × 94
INM-CM4-8	Institute for Numerical Mathematics, Russian Academy of Science, Moscow, Russia	180 × 120
INM-CM5-0		180 × 120
IPSL-CM6A-LR	Institute Pierre-Simon Laplace, France	144 × 143
KACE-1-0-G	National Institute of Meteorological Sciences, Korea Meteorological Administration, South Korea	192 × 144
MIROC6	University of Tokyo, Japan	256 × 128
MIROC-ES2L	University of Tokyo, Japan	128 × 64
MPI-ESM-1-2-HAM	Max Planck Institute, Germany	192 × 96
MPI-ESM1-2-HR		384 × 192
MPI-ESM1-2-LR		192 × 96
MRI-ESM2-0	Meteorological Research Institute, Japan Meteorological Administration, Japan	320 × 160
NESM3	Nanjing University of Information Science and Technology, China	192 × 96
NorCPM1	Norwegian Meteorological Institute, Oslo, Norway	144 × 96
SAM0-UNICON	Seoul National University, South Korea	288 × 192

Open in a new tab

b.
ERA5 reanalysis.

Hourly estimates of clouds and surface radiative fluxes from the fifth generation of European Center for Medium-range Weather Forecasts Reanalysis 5 (ERA5)¹³ are used in the study. The ERA5 reanalysis integrates observations from a wide variety of sources using enhanced modelling and data assimilation techniques, and provides data at global coverage with a horizontal resolution of 31 km on 137 levels from the surface up to 1 Pa.

c.
ARM observations.

The MOSAiC field campaign¹⁸ (2019/10/11–2020/10/01), conducted by the Department of Energy Atmospheric Radiation Measurement (ARM) program and its international partners, provides detailed process-level observations at fine scales to investigate the coupled central Arctic climate system. Here, hourly estimates of vertical profiles of clouds and surface downward longwave radiation from the ARM Best Estimate datasets³⁵(ARMBE) for MOSAiC are used. The ARMBE data for a Cloud Intensive Operational Period in March 2000 at the Southern Great Plains are also used to represent middle latitude conditions as a comparison.

The cloud fraction from ARMBE is assembled from the Active Remote Sensing of Clouds value-added product (VAP), which combines measurements from the Millimeter Wave Cloud Radars, Micropulse Lidars and laser ceilometers for cloud detection. The surface downward longwave radiation from ARMBE is assembled from the Data Quality Assessment for ARM Radiation Data VAP which provides the best estimate of surface radiative fluxes derived from radiometer systems.

Analysis methods

Net surface heat flux and cloud radiative effect.

The net surface heat flux ( Inline graphic ), defined as the total heat exchange between the atmosphere and the ocean, is calculated using the following Eq.

Here, Inline graphic , , and represents the net longwave radiation, net shortwave radiation, latent heat flux and sensible heat flux at the surface, respectively. Upward fluxes are considered as positive unless otherwise specified.

Another way to calculate Inline graphic is using the difference between the clear-sky radiation () and the cloud radiative effect (CRE)¹⁵ as shown below:

b.
Transmittance calculation.

The transmittance in Fig. 2 is calculated by using the Reference Forward Model (RFM)³⁶. The spectral absorption data are from the HITRAN2020. The water vapor continuum data is from the MT_CKD 4.1 model³⁷. Only the major isotopes of H₂O and CO₂ are considered. The calculation uses a resolution of 0.1 cm^− 1. It assumes a homogeneous path of 2 km thick at temperature of 245 K with 2.5 mm precipitable water for the Arctic and 296 K with 24 mm precipitable water for the US Standard Atmosphere at pressure of 1013.25 hPa. CO₂ concentration is assumed to be 400 ppmv.

c.
Normalization of clear-sky longwave radiation to the same temperature for Fig. 2c.

To remove the impact of surface temperature on the clear-sky radiation, we regressed Inline graphic against surface temperature (). The correlation coefficient is 0.93 with p < 0.01 (Fig. S1). The linear relationship is , where is in W/m² and is in K. The average temperature of the models (245 K) is then used to derive the normalized clear-sky radiative flux in all models. This clear-sky flux is then added to the cloud radiative effect and surface turbulent heat fluxes to obtain the normalized net surface heat fluxes in all models that are shown in Fig. 1c.

d.
Clear-sky radiation calculation.

The clear-sky radiative fluxes for the MOSAiC field and the March 2000 ARM SGP campaign shown in Figa. 3e and S2 are calculated by using the rapid radiative transfer model RRTMG^38,39. The model uses the correlated-k approach and it divides the infrared spectrum into 16 from spectral range from 3.1 μm to 1000.0 μm. The vertical profiles of temperature, water vapor and pressure are from the balloon soundings during the MOSAiC and SGP field campaigns processed as in the ARMBE. Concentrations of CO₂, O₃, N₂O, CH₄ and other trace gases are specified as in the Energy Exascale Earth System Model (E3SM) present-day climate simulations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1^{(487.6KB, docx)}

Acknowledgements

The authors sincerely appreciate the comments of the anonymous reviewers, which helped to improve the manuscript. This research was primarily supported by the Atmospheric Radiation Measurement (ARM) programs and partially supported by the Energy Exascale Earth System Model (E3SM) project, funded by the U.S. Department of Energy (DOE), Office of Science, Office of Biological and Environmental Research. Work at Lawrence Livermore National Laboratory (LLNL) was performed under the auspices of the U.S. DOE by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344. We wish to thank Dr. Anu Dudhia at the University of Oxford for making the Reference Forward Model available to us for this study.

Author contributions

C.T. and M.Z. conceived and designed the project. C.T. processed data from ERA5 reanalysis, ARM observations and CMIP6 simulations. M.Z. conducted the transmittance calculation. C.T. and M.Z. constructed the figures and tables and wrote the first draft. S.X. was responsible for acquiring the funding for the project and contributed to the interpretation of the results. All authors were involved in writing and editing subsequent drafts.

Data availability

ERA5 reanalysis data were downloaded from https://www.ecmwf.int/en/forecasts/dataset/ecmwf-reanalysis-v5. CMIP6 data are available via the Earth System Grid Federation at https://esgf-node.llnl.gov/search/cmip6/. Data from the ARM program can be accessed at the ARM Data Discovery at https://adc.arm.gov/discovery/.

Declarations

Competing interests

The authors declare no competing interests.

Footnotes

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Contributor Information

Cheng Tao, Email: tao4@llnl.gov.

Minghua Zhang, Email: minghua.zhang@stonybrook.edu.

References

1.Arias, P. A. Technical summary. in Climate Change: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change (eds Masson-Delmotte, V., Zhai, P., Pirani, A., Connors, S. L., Péan, C., Berger, S., Caud, N., Chen, L., Goldfarb, L., Gomis, M. I., Huang, M., Leitzell, K., Lonnoy, E., Matthews, J. B. R., Maycock, T. K., Waterfield, T., Yelekçi, O., Yu, R. & Zhou, B.) 33–44. (Cambridge University Press, 2021). 10.1017/9781009157896.002
2.Henke, M. et al. Assessment of Arctic Sea ice and surface climate conditions in nine CMIP6 climate models. Arct. Antarct. Alp. Res.55 (1). 10.1080/15230430.2023.2271592 (2023).
3.Davy, R. & Outten, S. The arctic surface climate in CMIP6: Status and developments since CMIP5. J. Clim.33, 8047–8068. 10.1175/JCLI-D-19-0990.1 (2020). [Google Scholar]
4.Notz, D. & Community, S. Arctic sea ice in CMIP6. Geophys. Res. Lett.47, e2019GL086749. 10.1029/2019GL086749 (2020). [Google Scholar]
5.Pan, R. et al. Future Arctic climate change in CMIP6 strikingly intensified by NEMO-family climate models. Geophys. Res. Lett.50, e2022GL102077. 10.1029/2022GL102077 (2023). [Google Scholar]
6.Screen, J. A. & Simmonds, I. The central role of diminishing sea ice in recent Arctic temperature amplification. Nature464, 1334–1337. 10.1038/nature09051 (2010). [DOI] [PubMed] [Google Scholar]
7.Smith, D. M. et al. Robust but weak winter atmospheric circulation response to future Arctic Sea ice loss. Nat. Commun.13, 727. 10.1038/s41467-022-28283-y (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Hong, Y. et al. Arctic-associated increased fluctuations of midlatitude winter temperature in the 1.5° and 2.0° warmer world. Npj Clim. Atmos. Sci.6, 26. 10.1038/s41612-023-00345-y (2023). [Google Scholar]
9.Ye, K. et al. Response of winter climate and extreme weather to projected Arctic sea-ice loss in very large-ensemble climate model simulations. Npj Clim. Atmos. Sci.7, 20. 10.1038/s41612-023-00562-5 (2024). [Google Scholar]
10.Bitz, C. M. & Lipscomb, W. H. An energy-conserving thermodynamic model of sea ice. J. Geophys. Res.104 (C7), 15669–15677. 10.1029/1999JC900100 (1999). [Google Scholar]
11.Eyring, V. et al. Overview of the coupled model intercomparison project phase 6 (CMIP6) experimental design and organization. Geosci. Model. Dev.9, 1937–1958. 10.5194/gmd-9-1937-2016 (2016). [Google Scholar]
12.Kacimi, S. & Kwok, R. Two decades of Arctic sea-Ice thickness from satellite altimeters: Retrieval approaches and record of changes (2003–2023). Remote Sens.16 (16), 2983. 10.3390/rs16162983 (2024). [Google Scholar]
13.Hersbach, H. et al. The ERA5 global reanalysis. Quart. J. Roy Meteorol. Soc.146, 1999–2049. 10.1002/qj.3803 (2020). [Google Scholar]
14.Tian, T. et al. Cooler Arctic surface temperatures simulated by climate models are closer to satellite-based data than the ERA5 reanalysis. Commun. Earth Environ.5, 111. 10.1038/s43247-024-01276-z (2024). [Google Scholar]
15.Ramanathan, V. et al. Cloud-radiative forcing and climate: Results from the earth radiation budget experiment. Science243, 57–63. 10.1126/science.243.4887.57 (1989). [DOI] [PubMed] [Google Scholar]
16.Standard Atmosphere, U. S., Government Printing, U. S. & Office, Washington, D. C. (1976). (1976). https://ntrs.nasa.gov/api/citations/19770009539/downloads/19770009539.pdf?attachment=true
17.Gordon, I. E. et al. The HITRAN2020 molecular spectroscopic database. J. Quant. Spectrosc. Radiat. Transf.277, 107949. 10.1016/j.jqsrt.2021.107949 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Shupe, M. et al. Overview of the MOSAiC expedition: Atmosphere. Elementa: Sci. Anthropocene10, 00060. 10.1525/elementa.2021.00060 (2022). [Google Scholar]
19.Cox, C. J. et al. Continuous observations of the surface energy budget and meteorology over the Arctic sea ice during MOSAiC. Sci. Data10, 519. 10.1038/s41597-023-02415-5 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Shupe, M. D., Matrosov, S. Y. & Uttal, T. Arctic mixed-phase cloud properties derived from surface-based sensors at SHEBA. J. Atmos. Sci.63, 697–711. 10.1175/JAS3659.1 (2006). [Google Scholar]
21.Morrison, H. et al. Resilience of persistent Arctic mixed-phase clouds. Nat. Geosci.5, 11–17. 10.1038/ngeo1332 (2012). [Google Scholar]
22.Persson, P. O. G. et al. Linking atmospheric synoptic transport cloud phase surface energy fluxes and sea-ice growth: Observations of midwinter SHEBA conditions. Clim. Dyn.49 (4), 1341–1364. 10.1007/s00382-016-3383-1 (2017). [Google Scholar]
23.Parker, C. L. et al. The influence of recent and future climate change on spring Arctic cyclones. Nat. Commun.13, 6514. 10.1038/s41467-022-34126-7 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Murto, S. L. et al. Extreme surface energy budget anomalies in the high Arctic in winter. J. Clim.36, 3591–3609. 10.1175/JCLI-D-22-0209.1 (2023). [Google Scholar]
25.Zhang, P. et al. More frequent atmospheric rivers slow the seasonal recovery of Arctic sea ice. Nat. Clim. Chang.13, 266–273. 10.1038/s41558-023-01599-3 (2023). [Google Scholar]
26.Shupe, M. D. & Intrieri, J. M. Cloud radiative forcing of the Arctic surface: The influence of cloud properties, surface albedo, and solar zenith angle. J. Clim.17, 616–628. https://doi.org/10.1175/1520-0442(2004)017<0616:CRFOTA>2.0.CO;2 (2004). [Google Scholar]
27.Becker, S., Ehrlich, A., Schäfer, M. & Wendisch, M. Airborne observations of the surface cloud radiative effect during different seasons over sea ice and open ocean in the Fram Strait. Atmos. Chem. Phys.23, 7015–7031. 10.5194/acp-23-7015-2023 (2023). [Google Scholar]
28.Xie, S., Zhang, M. et al. Simulations of midlatitude frontal clouds by single-column and cloud-resolving models during the atmospheric radiation measurement march 2000 cloud intensive operational period. J. Geophys. Res.11010.1029/2004JD005119 (2005). [Google Scholar]
29.Cess, R. D. et al. Interpretation of cloud-climate feedback as produced by 14 atmospheric general circulation models. Science245, 513–516. 10.1126/science.245.4917.513 (1989). [DOI] [PubMed] [Google Scholar]
30.Bony, S. et al. How well do we understand and evaluate climate change feedback processes? J. Clim.19, 3445–3482. 10.1175/JCLI3819.1 (2006). [Google Scholar]
31.Sherwood, S. C. et al. An assessment of Earth’s climate sensitivity using multiple lines of evidence. Rev. Geophys.58, e2019RG000678. 10.1029/2019RG000678 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
32.Solomon, A. et al. The winter central Arctic surface energy budget: a model evaluation using observations from the MOSAiC campaign. Elementa: Sci. Anthropocene11, 00104. 10.1525/elementa.2022.00104 (2023). [Google Scholar]
33.Wei, J., Ren, T., Yang, P., DiMarco, S. F. & Huang, X. Sensitivity of Arctic surface temperature to including a comprehensive ocean interior reflectance to the ocean surface albedo within the fully coupled CESM2. J. Adv. Model. Earth Syst.15, e2023MS003702. 10.1029/2023MS003702 [Google Scholar]
34.Seo, M., Kim, H. C., Seong, N. H., Sim, S. & Han, K. S. Variability of surface radiation budget over Arctic during two recent decades from perspective of CERES and ERA5 data. Remote Sens.15, 829. 10.3390/rs15030829 (2023). [Google Scholar]
35.Xie, S. et al. ARM climate modeling best estimate data: A new data product for climate studies. Bull. Amer Meteor. Soc.91, 13–20. 10.1175/2009BAMS2891.1 (2010). [Google Scholar]
36.Dudhia, A. The reference forward model (RFM). J. Quant. Spectrosc. Radiat. Transf.186, 243–253. 10.1016/j.jqsrt.2016.06.018 (2017). [Google Scholar]
37.Clough, S. A. et al. Atmospheric radiative transfer modeling: A summary of the AER codes. J. Quant. Spectrosc. Radiat. Transf.91, 233–244. 10.1016/j.jqsrt.2004.05.058 (2005). [Google Scholar]
38.Mlawer, E. J., Taubman, S. J., Brown, P. D., Iacono, M. J. & Clough, S. A. Radiative transfer for inhomogeneous atmospheres: RRTM, a validated correlated-k model for the longwave. J. Geophys. Res.102, 16663–16682. 10.1029/97JD00237 (1997). [Google Scholar]
39.Iacono, M. J., Mlawer, E. J., Clough, S. A. & Morcrette, J. J. Impact of an improved longwave radiation model, RRTM, on the energy budget and thermodynamic properties of the NCAR community climate mode, CCM3. J. Geophys. Res.105, 14873–14890. 10.1029/2000JD900091 (2000). [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Material 1^{(487.6KB, docx)}

Data Availability Statement

[CR1] 1.Arias, P. A. Technical summary. in Climate Change: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change (eds Masson-Delmotte, V., Zhai, P., Pirani, A., Connors, S. L., Péan, C., Berger, S., Caud, N., Chen, L., Goldfarb, L., Gomis, M. I., Huang, M., Leitzell, K., Lonnoy, E., Matthews, J. B. R., Maycock, T. K., Waterfield, T., Yelekçi, O., Yu, R. & Zhou, B.) 33–44. (Cambridge University Press, 2021). 10.1017/9781009157896.002

[CR2] 2.Henke, M. et al. Assessment of Arctic Sea ice and surface climate conditions in nine CMIP6 climate models. Arct. Antarct. Alp. Res.55 (1). 10.1080/15230430.2023.2271592 (2023).

[CR3] 3.Davy, R. & Outten, S. The arctic surface climate in CMIP6: Status and developments since CMIP5. J. Clim.33, 8047–8068. 10.1175/JCLI-D-19-0990.1 (2020). [Google Scholar]

[CR4] 4.Notz, D. & Community, S. Arctic sea ice in CMIP6. Geophys. Res. Lett.47, e2019GL086749. 10.1029/2019GL086749 (2020). [Google Scholar]

[CR5] 5.Pan, R. et al. Future Arctic climate change in CMIP6 strikingly intensified by NEMO-family climate models. Geophys. Res. Lett.50, e2022GL102077. 10.1029/2022GL102077 (2023). [Google Scholar]

[CR6] 6.Screen, J. A. & Simmonds, I. The central role of diminishing sea ice in recent Arctic temperature amplification. Nature464, 1334–1337. 10.1038/nature09051 (2010). [DOI] [PubMed] [Google Scholar]

[CR7] 7.Smith, D. M. et al. Robust but weak winter atmospheric circulation response to future Arctic Sea ice loss. Nat. Commun.13, 727. 10.1038/s41467-022-28283-y (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR8] 8.Hong, Y. et al. Arctic-associated increased fluctuations of midlatitude winter temperature in the 1.5° and 2.0° warmer world. Npj Clim. Atmos. Sci.6, 26. 10.1038/s41612-023-00345-y (2023). [Google Scholar]

[CR9] 9.Ye, K. et al. Response of winter climate and extreme weather to projected Arctic sea-ice loss in very large-ensemble climate model simulations. Npj Clim. Atmos. Sci.7, 20. 10.1038/s41612-023-00562-5 (2024). [Google Scholar]

[CR10] 10.Bitz, C. M. & Lipscomb, W. H. An energy-conserving thermodynamic model of sea ice. J. Geophys. Res.104 (C7), 15669–15677. 10.1029/1999JC900100 (1999). [Google Scholar]

[CR11] 11.Eyring, V. et al. Overview of the coupled model intercomparison project phase 6 (CMIP6) experimental design and organization. Geosci. Model. Dev.9, 1937–1958. 10.5194/gmd-9-1937-2016 (2016). [Google Scholar]

[CR12] 12.Kacimi, S. & Kwok, R. Two decades of Arctic sea-Ice thickness from satellite altimeters: Retrieval approaches and record of changes (2003–2023). Remote Sens.16 (16), 2983. 10.3390/rs16162983 (2024). [Google Scholar]

[CR13] 13.Hersbach, H. et al. The ERA5 global reanalysis. Quart. J. Roy Meteorol. Soc.146, 1999–2049. 10.1002/qj.3803 (2020). [Google Scholar]

[CR14] 14.Tian, T. et al. Cooler Arctic surface temperatures simulated by climate models are closer to satellite-based data than the ERA5 reanalysis. Commun. Earth Environ.5, 111. 10.1038/s43247-024-01276-z (2024). [Google Scholar]

[CR15] 15.Ramanathan, V. et al. Cloud-radiative forcing and climate: Results from the earth radiation budget experiment. Science243, 57–63. 10.1126/science.243.4887.57 (1989). [DOI] [PubMed] [Google Scholar]

[CR16] 16.Standard Atmosphere, U. S., Government Printing, U. S. & Office, Washington, D. C. (1976). (1976). https://ntrs.nasa.gov/api/citations/19770009539/downloads/19770009539.pdf?attachment=true

[CR17] 17.Gordon, I. E. et al. The HITRAN2020 molecular spectroscopic database. J. Quant. Spectrosc. Radiat. Transf.277, 107949. 10.1016/j.jqsrt.2021.107949 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR18] 18.Shupe, M. et al. Overview of the MOSAiC expedition: Atmosphere. Elementa: Sci. Anthropocene10, 00060. 10.1525/elementa.2021.00060 (2022). [Google Scholar]

[CR19] 19.Cox, C. J. et al. Continuous observations of the surface energy budget and meteorology over the Arctic sea ice during MOSAiC. Sci. Data10, 519. 10.1038/s41597-023-02415-5 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR20] 20.Shupe, M. D., Matrosov, S. Y. & Uttal, T. Arctic mixed-phase cloud properties derived from surface-based sensors at SHEBA. J. Atmos. Sci.63, 697–711. 10.1175/JAS3659.1 (2006). [Google Scholar]

[CR21] 21.Morrison, H. et al. Resilience of persistent Arctic mixed-phase clouds. Nat. Geosci.5, 11–17. 10.1038/ngeo1332 (2012). [Google Scholar]

[CR22] 22.Persson, P. O. G. et al. Linking atmospheric synoptic transport cloud phase surface energy fluxes and sea-ice growth: Observations of midwinter SHEBA conditions. Clim. Dyn.49 (4), 1341–1364. 10.1007/s00382-016-3383-1 (2017). [Google Scholar]

[CR23] 23.Parker, C. L. et al. The influence of recent and future climate change on spring Arctic cyclones. Nat. Commun.13, 6514. 10.1038/s41467-022-34126-7 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR24] 24.Murto, S. L. et al. Extreme surface energy budget anomalies in the high Arctic in winter. J. Clim.36, 3591–3609. 10.1175/JCLI-D-22-0209.1 (2023). [Google Scholar]

[CR25] 25.Zhang, P. et al. More frequent atmospheric rivers slow the seasonal recovery of Arctic sea ice. Nat. Clim. Chang.13, 266–273. 10.1038/s41558-023-01599-3 (2023). [Google Scholar]

[CR26] 26.Shupe, M. D. & Intrieri, J. M. Cloud radiative forcing of the Arctic surface: The influence of cloud properties, surface albedo, and solar zenith angle. J. Clim.17, 616–628. https://doi.org/10.1175/1520-0442(2004)017<0616:CRFOTA>2.0.CO;2 (2004). [Google Scholar]

[CR27] 27.Becker, S., Ehrlich, A., Schäfer, M. & Wendisch, M. Airborne observations of the surface cloud radiative effect during different seasons over sea ice and open ocean in the Fram Strait. Atmos. Chem. Phys.23, 7015–7031. 10.5194/acp-23-7015-2023 (2023). [Google Scholar]

[CR28] 28.Xie, S., Zhang, M. et al. Simulations of midlatitude frontal clouds by single-column and cloud-resolving models during the atmospheric radiation measurement march 2000 cloud intensive operational period. J. Geophys. Res.11010.1029/2004JD005119 (2005). [Google Scholar]

[CR29] 29.Cess, R. D. et al. Interpretation of cloud-climate feedback as produced by 14 atmospheric general circulation models. Science245, 513–516. 10.1126/science.245.4917.513 (1989). [DOI] [PubMed] [Google Scholar]

[CR30] 30.Bony, S. et al. How well do we understand and evaluate climate change feedback processes? J. Clim.19, 3445–3482. 10.1175/JCLI3819.1 (2006). [Google Scholar]

[CR31] 31.Sherwood, S. C. et al. An assessment of Earth’s climate sensitivity using multiple lines of evidence. Rev. Geophys.58, e2019RG000678. 10.1029/2019RG000678 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR32] 32.Solomon, A. et al. The winter central Arctic surface energy budget: a model evaluation using observations from the MOSAiC campaign. Elementa: Sci. Anthropocene11, 00104. 10.1525/elementa.2022.00104 (2023). [Google Scholar]

[CR33] 33.Wei, J., Ren, T., Yang, P., DiMarco, S. F. & Huang, X. Sensitivity of Arctic surface temperature to including a comprehensive ocean interior reflectance to the ocean surface albedo within the fully coupled CESM2. J. Adv. Model. Earth Syst.15, e2023MS003702. 10.1029/2023MS003702 [Google Scholar]

[CR34] 34.Seo, M., Kim, H. C., Seong, N. H., Sim, S. & Han, K. S. Variability of surface radiation budget over Arctic during two recent decades from perspective of CERES and ERA5 data. Remote Sens.15, 829. 10.3390/rs15030829 (2023). [Google Scholar]

[CR35] 35.Xie, S. et al. ARM climate modeling best estimate data: A new data product for climate studies. Bull. Amer Meteor. Soc.91, 13–20. 10.1175/2009BAMS2891.1 (2010). [Google Scholar]

[CR36] 36.Dudhia, A. The reference forward model (RFM). J. Quant. Spectrosc. Radiat. Transf.186, 243–253. 10.1016/j.jqsrt.2016.06.018 (2017). [Google Scholar]

[CR37] 37.Clough, S. A. et al. Atmospheric radiative transfer modeling: A summary of the AER codes. J. Quant. Spectrosc. Radiat. Transf.91, 233–244. 10.1016/j.jqsrt.2004.05.058 (2005). [Google Scholar]

[CR38] 38.Mlawer, E. J., Taubman, S. J., Brown, P. D., Iacono, M. J. & Clough, S. A. Radiative transfer for inhomogeneous atmospheres: RRTM, a validated correlated-k model for the longwave. J. Geophys. Res.102, 16663–16682. 10.1029/97JD00237 (1997). [Google Scholar]

[CR39] 39.Iacono, M. J., Mlawer, E. J., Clough, S. A. & Morcrette, J. J. Impact of an improved longwave radiation model, RRTM, on the energy budget and thermodynamic properties of the NCAR community climate mode, CCM3. J. Geophys. Res.105, 14873–14890. 10.1029/2000JD900091 (2000). [Google Scholar]

PERMALINK

Cloud radiative effect dominates variabilities of surface energy budget in the dark Arctic

Cheng Tao

Minghua Zhang

Shaocheng Xie