PIOMAS Variables on Model Grid

PIOMAS model grid data include model output for 1978-present. We provide the output from the Pan-Arctic Ice-Ocean Modeling and Assimilation System (PIOMAS) for best possible estimates of some key ice and ocean variables. The data sets only include results for the period of 1978-present when satellite ice concentration data are available for assimilation. These data sets include arctic sea ice thickness and concentration, snow depth, ice growth rate, ocean surface salinity, and more…

Data Access:

Access will be via “https” at the links provided in the table below or via sftp/rsync to sftp:anonymous©pscfiles•apl•uw•edu/pscfiles/zhang/PIOMAS/data/v2.1. with the following credentials:

Username: anonymous
Password: anonymous


The current primary version of PIOMAS is 2.1. 

A model integration using a slightly different configuration that only assimilates ice concentration at the ice edge but not in the interior of the sea ice pack is available at: sftp:anonymous©pscfiles•apl•uw•edu/pscfiles/zhang/PIOMAS/data/3sst

Format, grid etc. are the same as for version 2.1


VariableVariable name
in programs
DimensionUnitFile name
Sea ice thickness (Volume per unit Area), monthly mean    heff    2-D  mheff.H<yyyy>
Sea ice thickness in text files   heff.txt    2-D  mheff.txt<yyyy>
Sea ice thickness (Volume per unit Area) , daily mean    hiday    2-D  mhiday.H<yyyy>
Sea ice concentration    area    2-D area.H<yyyy>
Snow depth (water equivalent)    snow    2-D  msnow.H<yyyy>
Snow depth daily (water equivalent)   snowday  snowday.H<yyyy>
Sea ice growth or melt rate, (meter of ice)/s    iceprod    2-D  m/siceprod.H<yyyy>
Sea ice velocity, u and v components    icevel    2-D m/sicevel.H<yyyy>
Surface temperature (on snow/ice/ocean)    tice0    2-D  Ktice0.H<yyyy>
Sea ice advection (-delta(uh))   advect    2-D m/sadvect.H<yyyy>
Ocean heat flux used to melt ice, unit: (meter of ice)/s   oflux    2-D m/soflux.H<yyyy>
12-category ice thickness distribution    gice2Dsubgrid gice.H<yyyy>
Sea surface height     ssh    2-D cmssh.H<yyyy>
Ocean surface salinity (not exist)    osali1    2-D psuosali1.H<yyyy>
Ocean salinity of upper 10 levels    osali1_10    3-D psuosali1_10.H<yyyy>
Ocean surface temperature (not exist)    otemp1    2-D  Cotemp1.H<yyyy>
Ocean temperature of upper 10 levels   otemp1_10    3-D psuotemp1_10.H<yyyy>
Ocean velocity, upper 10 levels, u & v     uo1_10    2-D cm/suo1_10.H<yyyy>
Ocean velocity at 7.5 m depth, u & v, daily (not exist)     uo2    2-D cm/suo2day.H<yyyy>
P-E, NCEP/NCAR Reanalysis forcing  p-e.m.nmc   2-D m/sp-e.m.nmc.<yyyy>


* Ice growth rate is valid only in ice covered areas.

Methods and data:

SSM/I ice concentration data were assimilated in the TED sea ice model.


Grid Parameters:

Grid configurationGeneralized curvilinear coordinate system
Grid Size:360 x 120
Spatial Coverage45o N – 90o N
DomainThe grid domain is shown here
lon and lat for scalar fieldsgrid.dat
lon and lat for vector fieldsgrid.dat.pop
angle between latitude line
and  grid cell x-coordinate line


Temporal Coverage

Currently monthly data  from 1/1978 to 1/2017. Each file contains 12 monthly 2-D or 3-D fields.

Data Format:

The data are flat binary files (no headers) consisting of single-precision floating-point numbers. The data were written as unformatted direct access files on a Linux PC cluster. Byte order may have to be changed if you are using a machine using a different byte order (e.g. HP workstations ). A FORTRAN program demonstrating how to read all available variables and grid information is provided (read_360_120.f). A FORTRAN program demonstrating how to read sea ice thickness and velocity and grid information is also provided (read_hi_uice.f). A FORTRAN program shows how to calculate sea ice volume for various regions of the Arctic Ocean and subarctic seas (heff_for_volume.f). An IDL program demonstrating how to read scalar fields and grid information is also provided (read_draw_hi.pro). Another IDL routine (read_draw_uice.pro) is provided to read  vector fields and vector grid information. It also plots ice velocity on a spherical coordinate system using the angle between the latitude line and  the grid cell x-coordinate line (alpha.fortran.dat). The structure of the data sets should be apparent from these routines. Please let us know if  you have difficulty reading the data.


Fortran (yes it still exists):

read_360_120.f  : FORTRAN program  to read all available variables and grid information; note a need for some machines to use record numbering to read data
read_hi_uice.f : FORTRAN program  to read sea ice thickness and velocity and grid information, can be used to read other scalar and vector fields


read_draw_hi.pro: IDL program to read and plot sea ice thickness, can be used to read other scalar fields such as ice concentration and snow depth
read_draw_uice.pro : IDL program to read and plot sea ice velocity, can be used to read and plot other vector fields such as ocean velocity
colorbar.pro: IDL routine for drawing color bar
con_clr.pro: IDL routine for setting colors
redblue.pro: IDL routine for setting red and blue colors
LLtoXY.pro: IDL routine for plotting vectors
LLtoXY_s.pro: IDL routine for plotting vectors
XYtoLL.pro: IDL routine for plotting vectors
XYtoLL_s.pro: IDL routine for plotting vectors
heads.pro: IDL routine for plotting vectors
io.dat_360_120.output: model grid mask; ocean levels > 0, land <= 0
dz.dta30: Ocean level thickness (cm) (there are 30 ocean levels)

Python Tools (Community)

Check out Zack Labe’s Github repository of PIOMAS reading and plotting tools:

Robbie Mallet’s PIOMAS reader/netcdf converter:

Andrew Barret’s PIOMAS regridder:

Weiming Hu’s PIOMAS data downloader:


Let us know (schweig at uw edu) if you have things to share.

Data Access:

You can access the data via sftp/ssh/rsync protocols from ‘pscfiles.apl.washington.edu/pscfiles/zhang/PIOMAS/data/v2.1’, using the following credentials: username: ‘anonymous’ , password: ‘anonymous’. 

If you are downloading the data, please e-mail us at zhang©apl•washington•edu  (zhang©apl•washington•edu)   we
can update you on changes, bugs etc.


We’d appreciate your reference to the following paper for using the data:

Zhang, Jinlun and D.A. Rothrock: Modeling global sea ice with a thickness and enthalpy distribution model in generalized curvilinear coordinates, Mon. Wea. Rev. 131(5), 681-697, 2003.

Release Notes:

In February 2014 we identified a programming error affecting the  assimilation of ice concentration data. This error affected PIOMAS variables starting 2010-2013. Data have been reprocessed and are identified as version 2.1. Ice area and ice thicknesses in the Beaufort Chuckchi Sea areas were affected with ice thicknesses larger in reprocessed versions. Largest errors were in  May. Version 2. 0 data for changed years are available  (here).