Model data from: Revisiting the status of the Arctic’s Last Ice Area as a refuge for marine predators
Data files
May 05, 2026 version files 66.62 GB
-
ancil_data.tar
41.21 MB
-
CNRM-6-1-HR.tar
25.97 GB
-
ERA5.tar
14.72 GB
-
GFDL-ESM4.tar
25.89 GB
-
README.md
3.26 KB
Abstract
The Last Ice Area (LIA), located north of the Canadian Arctic Archipelago and Greenland, has the Arctic’s oldest and thickest sea ice. The LIA is hypothesized to be a potential climate refuge for ice-dependent top predators as Arctic sea ice continues to decline. However, recent studies indicate the region may be less resilient than expected. We used a coupled biophysical model to examine the impact of changes in sea ice and nutrient availability on the LIA marine planktonic ecosystem and the ecological potential of this critical area. The model was used to generate two downscaled simulations, which were forced by two IPCC AR6 climate models (GFDL-ESM4 and CNRM-6-1-HR using the SSP5-8.5 shared socioeconomic pathway), to investigate potential changes in primary productivity (PP) with different rates of future warming. Both downscaled model runs, which captured observed sea-ice dynamics, predicted declines in LIA sea-ice concentration and thickness. Under CNRM-6-1-HR, summer sea ice was largely absent from the LIA by 2055–2070, shortly after it disappeared elsewhere in the Arctic. Under GFDL-ESM4, some summer sea ice persisted through 2070, although concentration and thickness were low. Concurrently with declining sea ice, both models predicted increases in PP through 2070, with annual values peaking in August. While increased PP would support higher trophic levels, ice-dependent top predators require an ice platform for foraging and resting. Our approach of integrating physical and biological forecasts is a step towards a more complete picture to date of anticipated ecological changes in the LIA. We identify future research needs, which include evaluating additional climate models and forcing scenarios, collecting in-situ ecological data, including top predators, and a mechanistic understanding of how lower-level ecological changes will affect the life history of top predators.
Paper
Kristin L. Laidre, Axel Schweiger, Jinlun Zhang, Benjamin Cohen, Michael Steele, Steven H. Ferguson, G. W. K. Moore
“Revisiting the status of the Arctic’s Last Ice Area as a refuge for marine predators”, Ecosphere
Affiliations:
Laidre, Schweiger, Zhang, Cohen, Steele: Polar Science Center, Applied Physics Laboratory, University of Washington, Seattle, WA
Ferguson: Fisheries and Oceans Canada, Arctic Region, Winnipeg, MB, Canada
Moore: Department of Physics, University of Toronto, Canada
Contact
2026-03-30: Axel Schweiger (schweig@uw.edu)
Data Format
There are 4 separate data sets stored in 4 data files in tar format to maintain the directory structure. All files are in internally compressed netcdf-4 format so that additional compression will not likely result in significant reduction in file size.
This data set contains the numerical ice-ocean model (BIOMAS-L) output from the ERA-5, GFDL-ESM4 and CNRM-6-1-HR forced runs. The ERA-5 forced run is for 1979-2021, GFDL-ESM4 and CNRM-6-1-HR forced runs are from 1995-2070. BIOMAS-L model uses a curvilinear grid with variable horizontal grid spacing of about 12 km in the study area of the LIA but covers the entire Arctic at lower resolutions. sea ice fraction [sic] and ice thickness [sit] are provided as 3-dimensional (time,y,x) netcdf files at daily time resolution, with each file containing data for one year. Biological variables, small phytoplankton biomass [iSphy], large phytoplankton biomass [iLphy], small zooplankton biomass[iSzoo], large zooplankton biomass [iLzoo], predator plankton biomass [iPzoo], nitrate concentration [iNO3], ammonium concentration [iNH4], particulate organic nitrogen iPON], dissolved organic nitrogen [iDON], silicate concentration [iSiOH], particulate organic silicate [iopal], are provided at monthly time resolution as 4-dimensional variables (time,z,y,x). Where y (240) and x (752) are the horizontal dimensions, z (40) is the vertical/depth dimension. Time is provided in units of days since 1901-01-01. Ancillary data providing grids details including latitude, longitude of grid points, grid cell area, and land mask are in separate files.
Files/directory
- ancil_data.tar : ancil_data/
- static_scalar.nc : grid information for scalar files (ice thickness, ice concentration, biology)
- static_vector.nc : grid information for vector files (ice and ocean velocties)
- vertical_data.nc : vertical grid coordinates and mask
- ERA5.tar: ERA5 ERA5 data resulting from a hindcast using ERA5 atmospheric forcing (1979-2021)
- sit/hiday.HP[year].nc : daily ice thickness for [year]
- sic/aiday.H[year].nc : daily ice concentration for [year]
- bio/bio.H[year].nc : monthly biological variables for [year]
- GFDL-ESM4.tar: GFDL-ESM4 data resulting from a simulation using GFLD-ESM4 atmospheric forcing (1995-2070)
- (subdirectories as above)
- CNRM-6-1-HR.tar: CNRM-HR-1 data resulting from a simulation using CNRM-6-1-HR atmospheric forcing (1995-2070)
- (subdirectories as above)
