Contributors: L. Gilbert (University of Leeds), S. B. Simonsen (Technical University of Denmark), J. Wuite (ENVEO IT GmbH)
Issued by: University of Leeds / Lin Gilbert
Date: 02/12/2022
Ref: C3S2_312a_Lot4.WP1-PDDP-IS-v1_202206_IV_PQAD-v4_i1.1
Official reference number service contract: 2021/C3S2_312a_Lot4_EODC/SC1
History of modifications
List of datasets covered by this document
Related documents
Acronyms
General definitions
Ice Velocity: Ice flow velocity describes the rate and direction of ice movement. It is a fundamental parameter to characterize the behaviour of a glacier or an ice sheet. Ice velocity and its spatial derivative, strain rate (which is a measure of the ice deformation rate), are required for estimating ice discharge and mass balance and are essential input for glacier models that try to quantify ice dynamical processes.
Scope of the document
This document is the Product Quality Assurance Document (PQAD) for Ice Velocity (IV) as part of the Copernicus Ice Sheets and Ice Shelves service. The service addresses three essential climate variables (ECVs) by providing four separate products.
- Ice velocity is given for Greenland in product WP2-FDDP-IV-CDR
- Gravimetric mass balance is given for Greenland and Antarctica in product WP2-FDDP-GMB-CDR
- Surface elevation change is given for
- Antarctica in product WP2-FDDP-SEC-CDR-AntIS
- Greenland in product WP2-FDDP-SEC-CDR-GrIS
We document here the validation data, methods and results from the Climate Data Record (CDR) v3 for Ice Velocity. The same methods will be used to validate the v4 CDR.
Executive summary
In this document we describe the validation and quality assessment of the C3S Greenland Ice sheet ice velocity (IV) CDR. Chapter 1 details the IV products that are validated. These include the annual Greenland Ice Sheet ice velocity maps of 2017/18, 2018/19 and 2019/20. The products are derived from Sentinel-1 synthetic aperture radar (SAR) data using offset tracking techniques.
In Chapter 2 we describe the validation and intercomparison data sets used for the quality assessment. The quality assessment includes 1) detailed validation and intercomparison with contemporaneous validating datasets, these comprise in-situ Global Positioning System (GPS) data and publicly available ice velocity maps, and 2) testing of the algorithm performance in stable (ice-free) terrain.
The product validation methodology is described in Chapter 3. Detailed and component-wise intercomparisons with validation data sets are performed, and statistics (bias and standard deviation, Root Mean Square Error (RMSE)) are provided on the residuals and visualised in plots (e.g. scatterplots).
Chapter 4 provides a summary of the most recent (CDR v3) validation results. The intercomparison of ice velocity on selected outlet glaciers show a high level of agreement for the CDR v3 products, with a negligible overall mean bias and an RMSE of 0.18 m/d for the easting and 0.21 m/d for the northing component based for the 2017/18, 2018/19 and 2019/20 maps respectively. The ice sheet wide product intercomparison between Making Earth System Data Records for Use in Research Environments (MEaSUREs) and C3S indicate a negligible bias with an RMSE of 0.02-0-03 m/d. The outcome of the stable ground test indicates for all annual maps on average a negligible mean bias with an RMSE of 0.01-0-02 m/d for both easting and northing components.
1. Validated products
The ice velocity (IV) product assessment referred to in this document concerns the annually averaged IV maps of Greenland derived from Sentinel-1 Synthetic Aperture Radar (SAR) data acquired from 2017-10-01 to 2018-09-30, 2018-10-01 to 2019-09-30 (reprocessed from CDR v2) and 2019-10-01 to 2020-09-30 (combined as separate maps in CDR v3) (Figure 1.1, Table 1.1). The CDR v4 will extend the timeseries beyond the end of v3. The validation of CDR v4 will be described and discussed in the Product Quality Assessment Report [RD.1]. Surface velocity is derived by applying advanced iterative offset tracking techniques (RD.3). The ice velocity maps are annually averaged and provided at 250 m grid spacing (previously 500 m in CDR v2) in North Polar Stereographic projection (EPSG: 3413). The horizontal velocity is provided in true meters per day, towards easting (vx) and northing (vy) direction of the grid, and the vertical displacement (vz), is derived from a digital elevation model (TanDEM-X 90m digital elevation model (DEM); Rizzoli et al., 2017). The product is provided as a NetCDF4 file with the velocity components: vx, vy, vz and vv (magnitude of the horizontal components), along with maps showing the valid pixel count and uncertainty (std). For further details, see the Product User Guide and Specification (PUGS) document [RD.2]. For the product validation and intercomparisons discussed in this document, we consider the annually averaged maps as well as the individual (6/12-day repeat) ice velocity maps on which the annual maps are based. The individual maps (available through http://cryoportal.enveo.at1) are not provided as a separate product in C3S but are included here for validation as an extra quality assurance.
|
|
|
Figure 1.1: C3S ice velocity maps of the Greenland Ice Sheet based on Sentinel-1 data, 2017/18 (left), 2018/19 (middle), 2019/20 (right).
Table 1.1: Main characteristics of the C3S Greenland Ice Sheet ice velocity data sets.
Product | Temporal coverage | Spatial resolution |
|---|---|---|
CDR v3 2017-2018 Greenland IV | 2017-10-01 to 2018-09-30 | 250 m |
CDR v3 2018-2019 Greenland IV | 2018-10-01 to 2019-09-30 | 250 m |
CDR v3 2019-2020 Greenland IV | 2019-10-01 to 2020-09-30 | 250 m |
CDR v4 2020-2021 Greenland IV | 2020-10-01 to 2021-09-30 | 250 m |
2. Description of validating datasets
The quality assessment for ice velocity includes 1) detailed validation and intercomparison with contemporaneous validating datasets, these comprise in-situ GPS data and publicly available ice velocity maps, and 2) testing of the algorithm performance in stable (ice-free) terrain. Table 2.1 shows the main characteristics of the validation data.
Table 2.1: Main characteristics of validation data.
Validation Data | Program | Sensor | Spatial Coverage | Spatial resolution | Temporal Coverage | Temporal resolution |
|---|---|---|---|---|---|---|
In-situ GPS (v3) | PROMICE | GPS | AWS stations | point | 2007-2022 | hourly, daily, monthly, yearly |
TSX/TDX IV maps (v4) | MEaSUREs | TSX/TDX | major outlet glaciers | 100 m | 2008-2021 | 11 day |
Greenland IV mosaics (v3) | MEaSUREs | Sentinel-1, TSX/TDX, Landsat 8 | ice sheet wide | 200 m | 2015-2020 | 1 year |
Stable Terrain | Glacier-CCI | Landsat 7/8 | Greenland margins | 250 m | 1999-2004 | 5 year |
In-situ GPS data is available at various sites across the ice sheet (Figure 2.1). The GPS instruments are attached to Automatic Weather Stations (AWS) operated by the Geological Survey of Denmark and Greenland (GEUS) in collaboration with the National Space Institute at the Technical University of Denmark (DTU Space) and ASIAQ Greenland Survey as part of the Danish Programme for Monitoring of the Greenland Ice Sheet (PROMICE; Fausto and Van As, 2019). The GPS data is available through the PROMICE Data Portal2. The version 3 data set provides hourly, daily and monthly average positions and is currently updated until Jan 2022.
2 https://promice.org/ URL resource last accessed 2nd December 2022.
Figure 2.1: Locations of the PROMICE AWS stations equipped with GPS and used for ground validation.
The IV product is also evaluated against publicly available products covering the same area and time span. Although not a ground truth validation, such an inter-comparison provides a good level of quality assurance, particularly in areas where little change is to be expected. For product inter-comparison we utilize an independent collection of ice velocity maps derived from higher resolution TerraSAR-X/TanDEM-X (TSX/TDX) data, and covering the Greenland margins, as well as annual Greenland wide ice velocity maps, based on SAR and optical data. These IV maps were produced as part of the NASA 'Making Earth System Data Records for Use in Research Environments' (MEaSUREs) program (https://www.earthdata.nasa.gov/esds/competitive-programs/measures
).
The TerraSAR-X-based data set consists of a collection of IV maps covering the Greenland margins including most of the major outlet glaciers (Figure 2.2). The ice velocity is retrieved from repeat pass TerraSAR-X images (1 to 3 cycles) applying a combination of conventional InSAR and speckle tracking techniques (Joughin, 2002). For the validation assessment we use the latest version of the dataset (version 4) covering the time-period from June 2008 to Oct 2021. This dataset is available through the National Snow and Ice Data Center ( NSIDC) data portal (data set ID: NSIDC-0481; Joughin et al., 2021). Data files are delivered in GeoTIFF format at 100 m grid spacing in North Polar Stereographic projection (EPSG: 3413). Separate files are provided for the x and y velocity components along with corresponding error estimates for both velocity components.
|
|
Figure 2.2: Sentinel-1 IV mosaic showing locations of MEaSUREs TerraSAR-X derived IV maps used for product inter-comparison in black (left). The IV data sets cover the margins of Greenland and include most of the major outlet glaciers. The red circle indicates the location of C.H. Ostenfeld Glacier shown here as an example (right).
The Greenland-wide ice velocity maps used for product intercomparison are based on Sentinel-1, TerraSAR-X/TanDEM-X SAR images and Landsat-8 optical images. The maps are annually averaged from 1st December to 30th November and are provided at 200 m horizontal resolution in North Polar Stereographic projection (EPSG: 3413). For the validation assessment we use the latest version of the dataset, version 3, covering the time-period from 2015 to 2020. This dataset is available through the NSIDC data portal (data set ID: NSIDC-0725; Joughin, 2021).
Another method used here, for quality assessment of the ice velocity product, is the analysis of stable terrain, i.e. where no velocity is expected. This gives a good overall indication of the bias introduced by the end-to-end velocity retrieval including co-registration of images, velocity retrieval, etc. For masking the moving ice, we use a polygon shapefile of the ice sheet and peripheral glacier outlines produced by Rastner et al (2012, updated 2018). This shapefile is derived semi-automatically from Landsat 5 and 7, using a band ratio approach (red/SWIR) with scene specific thresholds and manual correction of debris cover, seasonal snow, shadow, water (outside of glaciers), sea ice and icebergs. By inverting the ice sheet/glacier shapefile, a land mask file is created as depicted in Figure 2.3.
Figure 2.3: Shapefile showing ice-free terrain (black) in Greenland and used for the stable terrain test.
3. Description of product validation methodology
Validation is done through detailed intercomparison of the IV product with in-situ GPS data and publicly available IV maps. As a measure of quality, we provide statistics on the component-wise (e.g for vx and vy) mean, the standard deviation and the root mean square error (RMSE) of the residuals (defined here as Sentinel-1 IV product minus validation dataset). Residuals larger than 1 m/d are excluded from the statistics as these contaminate the statistics and are in general obvious outliers that are easily filtered out in the IV products.Below follows specific information for each validation dataset.
1) Using GPS data: We use the monthly average positions provided in version 3 of the PROMISE AWS data set to calculate the local monthly averaged velocity magnitude, which is assigned to the station position. The most obvious outliers are removed manually. The monthly values are used to calculate yearly averages for Oct-Sept corresponding to the respective ice velocity map. For each monthly position, the corresponding pixel value in the Sentinel-1 IV map is selected and finally also averaged over a year. In this way there is one corresponding value for each station representing the annual mean IV. These are used to calculate the validation statistics and to create scatter plots for visualisation.
2) Inter-comparison with MEaSUREs TerraSAR-X-based data (data set ID: NSIDC-0481, Joughin et al., 2021): As a pre-processing step, the IV maps are first converted from m/y (metres per year) to m/d (metres per day) and bilinear resampled to a 250 m grid spacing. The separate vx and vy velocity files are then merged into a 2-band GeoTIFF to match the native Sentinel-1 IV maps geometry and format. Glacier velocity can fluctuate significantly over short time periods, particularly on the downstream sections of large outlet glaciers. Therefore, we only compare datasets derived from single Sentinel-1 repeat pass pairs (6/12 days only) and with a maximum of two days between the respective start dates. We intercompare both vx and vy components separately on a pixel-by-pixel basis and provide validation statistics for these.
3) Inter-comparison with MEaSUREs Greenland Ice Sheet Velocity Map (data set ID: NSIDC-0725, Joughin, 2021): Pre-processing is the same as described above. For the intercomparison, the 200 m product is resampled to 250 m with the same grid extent as the C3S product. The inter-comparison is performed on both the vx and vy components separately on a pixel-by-pixel basis. Validation statistics are provided for both.
4) Stable terrain test: The results for the ice covered (moving) area are separated from ice-free (stable) terrain. The masking is done using a polygon of the rock outlines derived from the icesheet/land/ocean boundaries. Validation statistics are provided for both velocity components.
4. Summary of most recent validation results
In this section, we provide a short summary of the CDR v3 validation results, which used the same analysis method as described above for CDR v4. The full CDR v4 validation and discussion will be published in related document, the Product Quality Assessment Report [RD.1].
Four different tests were performed for product quality assurance:
- Intercomparison of Sentinel-1 derived velocity with in-situ GPS measurements for 2017/18, 2018/19 and 2019/20.
- Intercomparison of individual 6-12-day repeat Sentinel-1 derived ice velocity maps, used to produce the annually averaged C3S ice velocity map, with contemporaneous MEaSUREs TSX derived ice velocity maps for selected major Greenland outlet glaciers that are covered by the validation dataset (Figure 2.2).
- Intercomparison of annually averaged Greenland Ice Sheet ice velocity map for 2017/18 and 2018/19 with MEaSUREs Greenland-wide IV maps.
- Stable terrain test, providing insight on the performance of the ice velocity retrieval algorithm, by analysing the results in stable terrain.
Table 4.1 provides a statistical overview of all inter-comparison results for CDR v3. For the annual maps the GPS intercomparison show excellent agreement with mean differences of only 0.01-0.02 m/d and an RMSE of 0.02-0.03 cm/d.
The intercomparison of ice velocity on selected outlet glaciers also show a high level of agreement for the CDR v3 products, with a negligible overall mean bias and an RMSE of 0.18 m/d for the easting and 0.21 m/d for the northing component based on sample sizes of 11.5 million, 10.4 million and 3.5 million for the 2017/18, 2018/19 and 2019/20 maps respectively.
For the ice sheet wide product intercomparison between MEaSUREs and C3S (only available for the 2017/18 and 2018/19 maps) the results indicate a negligible bias with an RMSE of 0.02-0-03 m/d.
Based on 5.1 million pixels the outcome of the stable ground test indicates for all annual maps on average a negligible mean bias with an RMSE of 0.01-0-02 m/d for both easting and northing components.
Table 4.1: Summary of the inter-comparison results for CDR v3. dmag, de and dn show the mean differences for the velocity magnitude, easting and northing, respectively. Similarly, RMSEmag, RMSEe and RMSEn show the root mean square error for velocity magnitude, easting and northing, respectively. Values are provided in m/d.
Product | Reference/Test | Pixels | dMag | RMSEMag | dE | RMSEE | dN | RMSEN |
|---|---|---|---|---|---|---|---|---|
2017/18 | GPS | 17 | 0.00 | 0.02 | - | - | - | - |
MEaSUREs TSX/TDX | 11.5 M | - | - | 0.00 | 0.18 | 0.00 | 0.21 | |
MEaSUREs (ice sheet) | 33.6 M | - | - | 0.00 | 0.02 | 0.00 | 0.03 | |
Stable Terrain | 5.1 M | - | - | 0.00 | 0.02 | 0.00 | 0.01 | |
2018/19 | GPS | 16 | 0.01 | 0.02 | - | - | - | - |
MEaSUREs TSX /TDX | 10.4 M | - | - | 0.00 | 0.18 | 0.00 | 0.22 | |
MEaSUREs (ice sheet) | 33.6 M | - | - | 0.00 | 0.03 | 0.00 | 0.03 | |
Stable Terrain | 5.1 M | - | - | 0.00 | 0.02 | 0.00 | 0.01 | |
2019/20 | GPS | 13 | 0.02 | 0.03 | - | - | - | - |
MEaSUREs TSX/TDX | 3.5 M | - | - | 0.00 | 0.17 | 0.00 | 0.20 | |
Stable Terrain | 5.1 M | - | - | 0.00 | 0.02 | 0.00 | 0.01 |
References
Fausto, R.S. and van As, D., (2019). Programme for monitoring of the Greenland ice sheet (PROMICE): Automatic weather station data. Version: v03, Dataset published via Geological Survey of Denmark and Greenland. DOI: https://doi.org/10.22008/promice/data/aws [URL resource last accessed 2nd December 2022].
Joughin, I. (2002). Ice-Sheet Velocity Mapping: A Combined Interferometric and Speckle-Tracking Approach. Annals of Glaciology 34: 195-201. doi: 10.3189/172756402781817978.
Joughin, I., Howat, I., Smith, B. and Scambos, T. (2021): MEaSUREs Greenland Ice Velocity: Selected Glacier Site Velocity Maps from InSAR, Version 4. Boulder, Colorado USA. NASA National Snow and Ice Data Center Distributed Active Archive Center.
Joughin, I. (2021): MEaSUREs Greenland Annual Ice Sheet Velocity Mosaics from SAR and Landsat, Version 3. Boulder, Colorado USA. NASA National Snow and Ice Data Center Distributed Active Archive Center.
Rastner, P., Bolch, T., Mölg, N., Machguth, H., Le Bris, R., Paul, F. (2012), updated 2018, The first complete inventory of the local glaciers and ice caps on Greenland. The Cryosphere, 6, 1483-1495. (doi:10.5194/tc-6-1483-2012)
Rizzoli, P., Martone, M., Gonzalez, C., Wecklich, C., Borla Tridon, D., Bräutigam, B., Bachmann, M., Schulze, D., Fritz, T., Huber, M., Wessel, B., Krieger, G., Zink, M., and Moreira, A. (2017): Generation and performance assessment of the global TanDEM-X digital elevation model. ISPRS Journal of Photogrammetry and Remote Sensing, Vol 132, pp. 119-139.
Acknowledgments of data contributors for IV validation
Data from the Programme for Monitoring of the Greenland Ice Sheet (PROMICE) and the Greenland Analogue Project (GAP) were provided by the Geological Survey of Denmark and Greenland (GEUS) at http://www.promice.dk. Data from the NASA Making Earth System Data Records for Use in Research Environments (MEaSUREs) Program were provided by the National Snow and Ice Data Center (NSIDC) at https://nsidc.org/data/measures.
