Methodology:

The first file, TopoR_icemask.csv, shows topographic-scale roughness, calculated from the variation in along-track topography (bed elevation measurements derived from the radar pulse) at a 200 m length-scale (the bin size used). This dataset presents the spatial pattern of roughness beneath the ice, clipped to ice mask, and filtered with respect to direction of flow. As this data is filtered with respect to flow, it allows for study of anisotropy in roughness (with regards to surface flow direction).

TopoR_icemask.csv includes data from all seasons from 1993 - 2016 campaigns. This dataset is derived using a moving 'bin' to calculate root mean square (rms) height for CReSIS along-track topography data (L2 RDS data). The flow filtering is done using Joughin et al (2016b) ice surface velocity mosaic.

The second file, ScatR_icemask.csv, shows scattering-derived roughness, calculated from quantifying characteristics of each bed-echo (the return from the radar pulse at the ice-bed interface). A roughness 'value' (expressed in 'metres') is inverted from the 'abruptness' (or pulse peakiness) of the bed return using a scattering model. Theoretically, this dataset presents the spatial pattern of roughness sensitive to roughness between the scale of the wavelength (order meter) and the Fresnel zone (order hundred meters) beneath the ice, clipped to ice mask. The 'scattering derived' roughness provides a more fine-scale roughness value (i.e. presents more information about roughness at scales less that 200 m).

ScatR_icemask.csv includes data from CReSIS seasons: 2011-2014 P3; 2011 TO, 2009 TO, 2008 TO, 2006 TO, 2005 TO, 2010 DC8. This dataset is calculated from bed-echo characteristics, using a scattering model, to invert rms height (scattering-derived roughness) from waveform abruptness; this is calculated using the CReSIS data (L1 RDS).

For the complete methodology, please see Cooper et al. (in prep.)

Data collection:

Data was collected with a Multichannel Coherent Radar Depth Sounder (MCoRDS) developed by the Center for Remote Sensing of Ice Sheets (CReSIS).

Data quality:

Quality control was taken from the original data source; only 'csv_good' files were used, and where appropriate 'high quality' picks, as per the CReSIS radar depth sounder documentation (see ftp://data.cresis.ku.edu/data/rds/rds_readme.pdf ). 'csv_good' files only contain data points where the ice surface and ice bottom were detected.

Both datasets are point data. For TopoR_icemask.csv, each point is the mid-point of a 200 m 'bin'.