More Like this

1. Brighter Ocean: Arctic sea ice solar partitioning

   https://doi.org/10.5061/dryad.n02v6wxb5  

   Webster, Melinda; Liu, Zheng; Light, Bonnie; Perovich, Donald (November 2025, Dryad)

   This dataset contains the partitioning of the surface downwelling radiative fluxes by the Arctic sea ice by reflection, absorption, and transmission. The dataset covers the period from 1984 to 2022 and was provided on 25 km resolution on the NSIDC EASE2 grid. # Brighter Ocean: Arctic sea ice solar partitioning Dataset DOI: [10.5061/dryad.n02v6wxb5](https://doi.org/10.5061/dryad.n02v6wxb5) ## Description of the data and file structure Our focus is on the large-scale partitioning of the incident solar radiation between reflection to the atmosphere (albedo, *a*), absorption in the ice (absorptance, *A*) and transmission to the ocean (transmittance, *T*). These parameters were determined every day from 1984 to 2024 at every point on the Equal Area Scalable Earth 2.0 (EASE2) grid, within the sea ice extent boundary. The basis for the analysis relies on several satellite and model products. These include geophysical variables of shortwave radiation, surface melt and freeze onset dates, as well as sea ice concentration, age, and thickness. Sea ice albedo is estimated using a multiphase albedo evolution determined for first year ice (Perovich and Polashenski, 2012) and multiyear ice (Perovich et al., 2002; Light et al., 2022). Distinct albedo evolutions for first-year and multiyear sea ice are computed using Version 4 of the EASE-grid Sea Ice Age product by Tschudi et al. (2019). The dates of melt and freeze-up book-end the Arctic melt season and govern the seasonal evolution of surface albedo of sea ice. For the onset dates of continuous melt and continuous freeze, we use passive microwave retrievals at 25-km resolution produced by Markus et al. (2009). The ice concentration is a key parameter in determining the solar partitioning. We use passive microwave retrievals from the NASA Bootstrap algorithm (Comiso, 1986), which are available in the NOAA/NSIDC Climate Data Record (Meier et al., 2021). The product is provided at daily 25-km gridded resolutions. In some instances, temporal gaps in the data exist due to inconsistent satellite coverage. Transmittance through the ice is defined by an exponential decay law of the form and corrected for the near infrared light absorbed by the snow and ice and not transmitted. The correction ratio is estimated to vary from about 0.55 for clear skies to 0.62 for complete cloud cover. An average value of 0.58 is used in this study. The ice extinction coefficient for visible light. Based on field observations (Light et al., 2008, 2015), we use 1.0 m-1 for an ice cover dominated by bare ice (0.5 < *a~i~* < 0.7), 3.0 m-1 for a cover dominated by snow (*a~i~* > 0.7), and 0.7 m-1 for an ice cover dominated by melt ponds (*a~i~* < 0.5). *S*ea ice thickness and is determined using the simulated thickness from the Pan-Arctic Ice Ocean Modeling and Assimilation System (PIOMAS) (Zhang and Rothrock, 2003). ### Files and variables #### File: BO_EASE2_gridded_25km_YYYY.nc **Description:** for year YYYY. **Variables:**  dimensions: X = 263, Y = 263, time=365 or 366. insol (time,Y,X): surface downwelling solar radiative flux sic (time,Y,X): sea ice concentration sith (time,Y,X): sea ice thickness iceage (time,Y,X): sea ice age emelt, efreeze, fmelt, ffreeze (Y,X): Early ('e')/full ('f') melt/freeze onset date, in day-of-year. alb_yyy (time,Y,X): ice albedo using albedo scheme yyy. The albedo scheme include MYI (using multi-year ice scheme), FYI (using first-year ice scheme), and AGE (combined multi-year and first-year albedo depending on ice age). h_xxx_yyy (time,Y,X): heat flux on xxx surface type using albedo scheme yyy, with unit of W/m^2^. The xxx surface types include nIC (assuming fully ice covered), ice (ice portion of the gridbox), ocn (open water portion of the gridbox). The yyy albedo scheme include MYI (using multi-year ice scheme), FYI (using first-year ice scheme), and AGE (combined multi-year and first-year albedo depending on ice age). An exemption is h_ocn_SIC, the heat input to the open water portion of the gridbox using the sea ice concentration. accu_h_xxx_yyy (time,Y,X): the accumulated heat input into the gridbox, with unit of MJ/m^2^. hthr_ice_xxx_KAPPA0 (time,Y,X): the heat through ice portion of the gridbox, using albedo scheme xxx and constant ice extinction coefficient kappa of 1, with unit of W/m^2^. hthr_ice_xxx_KAPPA_ALB (time,Y,X): the heat through ice portion of the gridbox, using albedo scheme xxx and ice extinction coefficient kappa depending on ice albedo value, with unit of W/m^2^. accu_hthr_ice_xxx_KAPPA0 (time,Y,X): the accumulated heat through ice portion of the gridbox, using albedo scheme xxx and constant ice extinction coefficient kappa of 1, with unit of MJ/m^2^. accu_hthr_ice_xxx_KAPPA_ALB (time,Y,X): the accumulated heat through ice portion of the gridbox, using albedo scheme xxx and ice extinction coefficient kappa depending on ice albedo value, with unit of MJ/m^2^. ## Code/software The software for processing and analyze this dataset is published on Zenodo: [10.5281/zenodo.17675721](https://doi.org/10.5281/zenodo.17675721) The developmental package is hosted o GitHub: [https://github.com/liuzheng-arctic/BrighterOcean](https://github.com/liuzheng-arctic/BrighterOcean) 

   more » « less
2. Daily CMIP6 and NSIDC CDR (National Snow and Ice Data Center Climate Data Record) Arctic sea ice area and sea ice extent, 1980-2100

   https://doi.org/10.18739/A2CC0TV9V  

   Jahn, Alexandra; Heuzé, Céline (January 2024, NSF Arctic Data Center)

   This dataset contains the daily Arctic sea ice area (SIA) and sea ice extent (SIE) data for all CMIP6 models and the historical period based on the NOAA/NSIDC Climate Data Record (CDR) created for Heuzé and Jahn, The first ice-free day in the Arctic Ocean could occur before 2030, accepted, Nature Communications. This is a derived dataset based on publicly available underlying data: - For the CMIP6 data, the SIA and SIE data included here is based on the daily siconc and siconca CMIP6 model output freely available on the CMIP6 data portals (https://pcmdi.llnl.gov/CMIP6/). These pan-Arctic daily SIA and SIE were calculated north of 30N, on each model's native grid, using each models grid area data (areacello or areacella). SIA was defined as sea ice concentration multiplied by the grid cell area and summed over all grid cells. SIE was defined as the sum of the grid cell area for all grid cells where the sea ice concentration was larger than 0.15. All processed SIA and SIE data is included in this dataset, even if the model was later excluded from the analysis for one reason or another (see Heuzé and Jahn 2024, Methods section). All data included has the same number of days as the underlying model. The historical data spans 1980-2014 and can be found in the CMIP6_historical_data.zip file, and the scenario data spans 2015 to the end of the 21st century simulation, for multiple scenarios (SSPs), and can be found in CMIP6_ssp_data.zip. Files are provided as .zip files to make it easy to download all data at once, as the SIA and SIE data is saved in one file per model and ensemble member, and for the scenario simulations, also per ssp. - For the NOAA/NSIDC Climate Data Record (CDR), the SIA and SIE data included here is based on the NOAA/NSIDC Climate Data Record of Passive Microwave Sea Ice Concentration, Version 4, doi:10.7265/efmz-2t65, Meier et al 2021. The sea ice concentration is multiplied by the grid size of each grid box, for this data, 25x25 kilometers (km) = 625 kilometers squared (km2), and then summed over the full domain. In doing that, we include the interpolated data in the pole hole as included in the sea ice concentration data, but exclude all land/coastal grid points (i.e., values &gt; 2.5 in the underlying data). As the filename indicates, we removed all leap year data from this data (dropped every Feb 29th) so that all years have 365 days. Note that while the file name says this data is for 19790101 to 20231231, it does indeed include 1978 as first year (so 1978-01-01-2023-12-31), with daily data starting on 1978-10-25 (nan before then). We did not change the name of the data file to still allow all archived scripts using this datafile to run. Scripts that work on this data associated with Heuzé and Jahn (2024) can be found at: https://zenodo.org/records/14008665, doi:10.5281/zenodo.14006059 References: Meier, W. N., F. Fetterer, A. K. Windnagel, and S. Stewart. 2021. NOAA/NSIDC Climate Data Record of Passive Microwave Sea Ice Concentration, Version 4. Boulder, Colorado, USA. NSIDC: National Snow and Ice Data Center https://doi.org/10.7265/efmz-2t65 

   more » « less
3. Interannual variability of Arctic sea ice concentration and area for six climate model large ensembles and four observational datasets, 1979-2020

   https://doi.org/10.18739/A2H98ZF3T  

   Wyburn-Powell, Christopher (January 2022, NSF Arctic Data Center)

   This dataset includes statistically resampled monthly time series data of Arctic sea ice area and gridded data for March and September for sea ice concentration for a selection of large ensemble climate models and observational datasets. Arctic sea ice concentrations and areas are resampled from all available members of six coupled climate models from the Coupled Model Intercomparison Project 5 (CMIP5). These six models are: The second generation Canadian Earth System Model (CanESM2), The Community Earth System Mode version 1 (CESM1), The Commonwealth Scientific and Industrial Research Organisation Global Climate Model Mark 3.6 (CSIRO MK3.6), The Geophysical Fluid Dynamics Laboratory Coupled Climate Model version 3 (GFDL CM3), Geophysical Fluid Dynamics Laboratory Earth System Model version 2 with Modular Ocean Model version 4.1 (GFDL ESM2M), Max Planck Institute Earth System Model version 1 (MPI ESM1). The Four observational datasets are The Hadley Centre Sea Ice and Sea Surface Temperature data set version 1 (HadISST1), The National Oceanic and Atmospheric Administration and National Snow and Ice Data Center Climate Data Record Version 4 (CDR), The The National Aeronautics and Space Administration Team Algorithm (NT), and the The National Aeronautics and Space Administration Bootstrap Team Algorithm (BT). The sea ice area data is resampled 10,000 times and then the standard deviation of those resamplings is calculated, which can be considered analagous to interannual variability of sea ice area (SIA). The standard deviation (sigma) and mean (mu) of these data represent the variability and typical values respectively of interannual variability found in each ensemble member or observational dataset. Sea ice concentration is resampled 1000 times with the same standard deviation and mean metrics for sea ice concentration. This dataset was created to evaluate climate model projections of Arctic sea ice interannual variability and is used in the article Wyburn-Powell, Jahn, England (2022), Modeled Interannual Variability of Arctic Sea Ice Cover is Within Observational Uncertainty, Journal of Climate, https://doi.org/10.1175/JCLI-D-21-0958.1. This work was conducted at the University of Colorado Boulder from 2020-2022. The figures from the Journal of Climate article can be reproduced from the following datasets. The code used to create the datasets can be located at https://www.doi.org/10.5281/zenodo.6687725. - Figure 1: Sigma_obs_SIA.nc - Figure 2: Sigma_obs_SIA.nc, Mu_obs_SIA.nc, Sigma_mem_SIA.nc, Mu_mem_SIA.nc - Figure 3: Sigma_mem_varying_time_periods_1965_2066_03.nc, Sigma_LE_varying_time_periods_1965_2066_03.nc, Sigma_LE_varying_time_periods_1970_2040_09.nc, Sigma_obs_varying_time_periods_1953_2020.nc - Figure 4: Sigma_obs_SIA.nc, Sigma_mem_SIA.nc - Figure 5: Sigma_obs_SIA.nc - Figure 6: &lt;model_name&gt;_resampled_0&lt;month&gt;_individual.nc, &lt;observational_dataset&gt;_resampled_individual_1979_2020_03_09.nc - Figure 7: Sigma_obs_SIA.nc, Mu_obs_SIA.nc, Sigma_mem_SIA.nc, Mu_mem_SIA.nc - Figure 8: &lt;model_name&gt;_resampled_0&lt;month&gt;_individual.nc, &lt;observational_dataset&gt;_resampled_individual_1979_2020_03_09.nc - Figure 9: Sigma_mem_SIA.nc, Sigma_LE_SIA.nc 

   more » « less
4. Predictability of the Minimum Sea Ice Extent from Winter Fram Strait Ice Area Export: Model versus Observations

   https://doi.org/10.1175/JCLI-D-22-0931.1  

   Trotechaud, Sandrine; Tremblay, Bruno; Williams, James; Romanski, Joy; Romanou, Anastasia; Bushuk, Mitchell; Merryfield, William; Msadek, Rym (April 2024, Journal of Climate)

   Observations show predictive skill of the minimum sea ice extent (Min SIE) from late winter anomalous offshore ice drift along the Eurasian coastline, leading to local ice thickness anomalies at the onset of the melt season—a signal then amplified by the ice–albedo feedback. We assess whether the observed seasonal predictability of September sea ice extent (Sept SIE) from Fram Strait Ice Area Export (FSIAE; a proxy for Eurasian coastal divergence) is present in global climate model (GCM) large ensembles, namely the CESM2-LE, GISS-E2.1-G, FLOR-LE, CNRM-CM6-1, and CanESM5. All models show distinct periods where winter FSIAE anomalies are negatively correlated with the May sea ice thickness (May SIT) anomalies along the Eurasian coastline, and the following Sept Arctic SIE, as in observations. Counterintuitively, several models show occasional periods where winter FSIAE anomalies are positively correlated with the following Sept SIE anomalies when the mean ice thickness is large, or late in the simulation when the sea ice is thin, and/or when internal variability increases. More important, periods with weak correlation between winter FSIAE and the following Sept SIE dominate, suggesting that summer melt processes generally dominate over late-winter preconditioning and May SIT anomalies. In general, we find that the coupling between the winter FSIAE and ice thickness anomalies along the Eurasian coastline at the onset of the melt season is a ubiquitous feature of GCMs and that the relationship with the following Sept SIE is dependent on the mean Arctic sea ice thickness. 

   more » « less
5. Benchmark seasonal prediction skill estimates based on regional indices

   https://doi.org/10.5194/tc-13-1073-2019  

   Walsh, John E.; Stewart, J. Scott; Fetterer, Florence (January 2019, The Cryosphere)

   Abstract. Basic statistical metrics such as autocorrelations and across-region lagcorrelations of sea ice variations provide benchmarks for the assessments offorecast skill achieved by other methods such as more sophisticatedstatistical formulations, numerical models, and heuristic approaches. In thisstudy we use observational data to evaluate the contribution of the trend tothe skill of persistence-based statistical forecasts of monthly and seasonalice extent on the pan-Arctic and regional scales. We focus on the BeaufortSea for which the Barnett Severity Index provides a metric of historicalvariations in ice conditions over the summer shipping season. The varianceabout the trend line differs little among various methods of detrending(piecewise linear, quadratic, cubic, exponential). Application of thepiecewise linear trend calculation indicates an acceleration of the winterand summer trends during the 1990s. Persistence-based statistical forecastsof the Barnett Severity Index as well as September pan-Arctic ice extent showsignificant statistical skill out to several seasons when the data includethe trend. However, this apparent skill largely vanishes when the data aredetrended. In only a few regions does September ice extent correlatesignificantly with antecedent ice anomalies in the same region more than 2months earlier. The springtime “predictability barrier” in regionalforecasts based on persistence of ice extent anomalies is not reduced by theinclusion of several decades of pre-satellite data. No region showssignificant correlation with the detrended September pan-Arctic ice extent atlead times greater than a month or two; the concurrent correlations arestrongest with the East Siberian Sea. The Beaufort Sea's ice extent as farback as July explains about 20 % of the variance of the Barnett SeverityIndex, which is primarily a September metric. The Chukchi Sea is the onlyother region showing a significant association with the Barnett SeverityIndex, although only at a lead time of a month or two. 

   more » « less