UPDATE: 2021
The data has now been updated to cover the full CryoSat-2 time period and spatial coverage. There is an updated NetCDF file format for the new data.
For the Arctic we use a 20 km gird on a Polar Stereographic projection.
For the Antarctic we use a 50 km gird on a Polar Stereographic projection. To rotate the ocean velocities to be square with lon/lat use the equation:
Ulat = Geo_surface_current_y*ang_c
+ Geo_surface_current_x*ang_s
Ulon = Geo_surface_current_x*ang_c
- Geo_surface_current_y*ang_s
Download the new data files after registering at the bottom of this page.
Data Description of Original DOT Product (Armitage)
A full description of the data processing and uncertainties is given by Armitage et al. (2016).
The data consists of monthly fields of dynamic ocean topography (DOT) and geostrophic currents covering 60N to 81.5N, for the period 2003-2014. Envisat data are used for 2003-2011 and CryoSat-2 data are used for 2012-2014. In sea ice-covered regions of the Arctic, SSH is estimated from data originating from leads or polynyas and this has been combined with regular open ocean estimates of sea surface height. Data from the two satellites have been cross-calibrated in the mission overlap period (November 2011 - March 2012) and agree within 1.0cm RMS difference, and correlation of 0.94.
Sea surface height estimates are referenced to the GOCO03s combined satellite-only geoid model (http://www.goco.eu/). The data are averaged in 0.75° x longitude x 0.25° latitude grid cells and grid cells are masked if they are within 10km of land to avoid issues with land contamination. The resulting grid is then smoothed with a Gaussian convolution filter, with a standard deviation of 100km and a radius of 3 standard deviations. DOT is in units of meters; geostrophic currents are in meters per second. The data is stored in netCDF format, with the following properties:
The data is set to NaN over land and in ocean regions that fall within the land mask. There are some residual issues with the way that the orbit pattern of CryoSat-2 samples the ocean surface throughout a given month, which leads to some striping in the data (see, for example, the Siberian shelf seas in November 2012). These effects should be kept in mind when using the data to estimate geostrophic currents, however they are generally quite localised, normally only affecting the Siberian shelf seas, and appear fairly infrequently. Note also that this does not necessarily constitute an error in the CryoSat-2 data, but is a result of the spatio-temporal sampling of the time varying DOT over the course of a month.
Reference
If you use the DOT/geostrophic currents data (delete as appropriate), please include the following acknowledgment and reference in your work:
Arctic dynamic topography/geostrophic currents data were provided by the Centre for Polar Observation and Modelling, University College London (www.cpom.ucl.ac.uk/dynamic_topography) [Armitage et al. (2016)/Armitage et al. (2017)]
Armitage,T.W.K., S.Bacon, A.L.Ridout, S.F.Thomas, Y.Aksenov and D.J.Wingham (2016), Arctic sea surface height variability and change from satellite radar altimetry and GRACE, 2003-2014, Journal of Geophysical Research:Oceans, 121, doi: 10.1002/2015JC011579
Armitage, T.W.K., S.Bacon, A.L.Ridout, A.A.Petty, S.Wolbach and M.Tsamados (2017), Arctic Ocean surface geostrophic circulation 2003-2014, The Cryosphere, 11, doi: 10.5194/tc-11-1767-2017
Data Access
To access the data please provide us with your contact details so that we can let you know about any future updates.