Last modified on Apr 08, 2024 12:04

Contributors: University of Zurich: Jacqueline Bannwart, Inés Dussailant, Frank Paul, Michael Zemp

Issued by: UZH/Frank Paul, Jacqueline Bannwart

Date: 16/08/2023

Ref: C3S2_312a_Lot4.WP2-FDDP-GL-v1_202212_A_PUGS-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

Brokered data set: A dataset that is made available in the Climate Data Store (CDS) but freely available (under given license conditions) from external sources. In the case of the glacier distribution service, the Randolph Glacier Inventory (RGI) is brokered for the CDS from https://glims.org/RGI under a CC-BY 4.0 license.

Crevasse: A crevasse is a wedge-shaped crack in the surface of a glacier caused by differential movement (faster towards the centre, slower towards the boundary) of the ice. A crevasse can be several metres wide and more than 30 m deep.

Debris-cover: Debris on a glacier is usually composed of unsorted rock fragments with highly variable grain size (from mm to several m). These might cover the ice in lines of variable width separating ice with origin in different accumulation regions of a glacier (so called medial moraines) up to a complete coverage of the ablation region. Automated mapping of glacier ice is only possible when the debris is not covering the ice completely when compared to the image pixel size.

Glacier complex: A contiguous ice mass that is the result of the binary (yes/no) glacier classification after conversion from raster to vector format. Usually the glacier complexes are divided into individual glaciers by digital intersection with a vector layer containing hydrologic divides derived from watershed analysis of a digital elevation model (DEM).

Hypsometry: The hypsometry of a glacier gives the distribution of area with elevation, i.e. for each elevation interval covered by a glacier the area in this interval is listed.

Scope of the document

This document is the Product User Guide and Specification (PUGS) for the glacier area (outlines) product from the glacier distribution service. It describes the Climate Data Records (CDRs) brokered to the CDS from the Randolph Glacier Inventory (RGI) version 6.0 in its vector (shape file) and raster (netCDF) formats. For the detailed product description and usage, we refer to the freely available reference documents [RD1] and [RD2].

Executive summary

This document provides a description of the Climate Data Record (CDR) dataset provided by the Copernicus Climate Change Service (C3S) Glacier Distribution Service to the Climate Data Store (CDS): the RGI containing glacier outlines (shapefile format) with glacier-specific attribute information and hypsometric information (csv file) for the c. 215,000 glaciers on Earth. In this updated version we also describe details of the related raster dataset that has been derived from the RGI.

In Section 1, we first provide a description of the datasets and methods used to create the RGI, before we give an overview on its spatio-temporal coverage and usage limitations. Details of the technical specifications and data fields are described in the following Section 2, including some illustrations visualizing the information provided. Information on data access is provided in Section 3.

1. The Randolph Glacier Inventory (RGI)

1.1. Product description

The RGI is a globally almost complete collection of digital glacier outlines that was compiled by the science community in a short period of time to serve as a base for global-scale glacier-related calculations in support of Intergovernmental Panel on Climate Change Fifth Assessment Report (IPCC AR5) (Vaughan et al. 2013). It is largely based on glacier outlines from the Global Land Ice Measurements from Space (GLIMS) database (glims.org) and additional outlines derived from mostly Landsat but also Advanced Spaceborne Thermal Emission and reflection Radiometer (ASTER) satellite images. In a few regions the outlines are also based on digitised topographic maps, or nominal data from the World Glacier Inventory (WGI) are used and glaciers are represented by circles of the related size (Pfeffer et al. 2014). Outlines have been derived from the satellite images by a range of methods, from fully manual digitisation to automated methods to map clean ice (e.g. band ratio) and manual editing to include omission (debris cover, clouds, ice in shadow) and exclude commission (lakes, seasonal snow, rock in shadow) errors. Hence, all glacier outlines are quality checked against a reference dataset but the applied corrections are based on the experience of the analyst and interpretation differences exist (see Product Quality Assurance Document (PQAD), RD3). A Digital Elevation Model (DEM) is the key auxiliary dataset used to obtain (a) drainage divides (e.g. Bolch et al. 2010, Kienholz et al. 2013) to separate glacier complexes into individual glaciers, and (b) calculate topographic information (e.g. min, max, mean, median elevation, slope and aspect, hypsometry) for each glacier (e.g. Paul et al. 2002 and 2009). As a range of different DEMs (e.g. ASTER Global Digital Elevation Model (GDEM), Shuttle Radar Topography Mission (SRTM), National Elevation Dataset (NED)) have been used to derive both (a) and (b), there is also some variability in the position of joint glacier boundaries and topographic parameters (Frey and Paul 2012). In any case, the derived glacier outlines are categorical variables representing a Level 3 Earth Observation (EO) data processing.

In the present document there will be no details regarding the individual datasets used for the RGI. Instead they can be found in the corresponding description of these datasets in the RGI Technical Note [RD1] that is available online. This document lists for all datasets the regions covered, its contributors, the satellite data and processing methods used, and the history of updates since RGI 1.0. A summarised description of the RGI and its contents can be found in the freely available study by Pfeffer et al. (2014, see  [RD2] for the DOI). This overview paper provides also details on the RGI history, the merging of the contributing datasets, the applied measures for quality control, and known limitations of the dataset.

1.2. Target requirements

The RGI covers all regions with glaciers globally. To ease data handling, the entire dataset has been subdivided into 19 regions (Figure 1). The temporal coverage is highly variable but clusters around the years 2000 and 2010 (Figure 2). Random and systematic errors have not been determined regularly in the related studies, but it can be noted that systematic errors (omission and commission, see Section 1.1) are removed by the analyst during manual editing. The remaining random errors are related to a wide range of sources (interpretation, geolocation, pixel size, contrast) and a range of methods have been applied to assess them. A condensed overview of the error terms and a generalised assessment for the datasets in the RGI is presented in [RD2] and the PQAD [RD3]. In the following, we provide a short description of the fields in the Global Climate Observing System (GCOS) requirements for the Essential Climate Variable (ECV) ‘Glaciers’ (Table 1) that is taken from GCOS (2016) (see also RD4).

Figure 1: Glacier coverage in the RGI (red) and the 19 main regions (black squares). The image is taken from Pfeffer et al. (2014).

Figure 2: Temporal coverage in the RGI 4.0 (red line) and 5.0 (orange bars) in relation to glacier count (top) and area covered (bottom). The image is taken from RGI Consortium (2017).

Glacier area is the map-projected size of a glacier in km². Typically, a minimum size of 0.01 or 0.02 km² is applied, as a glacier must flow by definition and ice patches smaller than this are likely stagnant. The frequency “Annual” means that each year the availability of satellite (or aerial) images should be checked to not miss any images acquired in the year with the best snow conditions (i.e. snow is only found on glaciers and not hiding their true extent). However, the update cycle of glacier inventories should be a few decades (the typical response time of glaciers) according to Global Terrestrial Network for Glaciers (GTN-G). The horizontal resolution of 15-30 m refers to typically used satellite sensors (Landsat and ASTER) to map glaciers. At coarser resolution the quality of the derived outlines rapidly degrades. Spatial resolutions better than 15 m (e.g. the 10 m from Sentinel-2) are preferable, as typical characteristics of glacier flow (e.g. crevasses) only become visible at this resolution (Paul et al. 2016). The measurement uncertainty refers to the accuracy of the glacier area. Although for clean ice this value can be reached using automated methods, for debris-covered glaciers manual editing is required to have a product accuracy better than this (Paul et al. 2013).

Table 1: GCOS requirements for the ECV Glaciers. Please note, the cited 'IGOS (2009)' document is IGOS (2007) in the reference list.

1.3. Data usage information

The technical details about the RGI (e.g. data format, file names, product content, attributes, quality indicators, projection etc.) are described in Section 2. Usage restrictions can be found in [RD2]. Key points to consider are:

• The RGI is a vector dataset in shape file format consisting of about 215 000 glaciers with related attribute information and comes in a geographic projection with WGS1984 datum.
• The RGI has been created for continental to global-scale applications and might, thus, have regional or local quality issues. One goal of the C3S glacier distribution service is to improve on the latter.
• Whereas the GLIMS database (from which the RGI is derived) is multi-temporal and includes all datasets provided by the community, the RGI is a frozen snapshot including only one outline from each glacier.
• It is not recommended to use the RGI for change assessment without carefully checking its quality in the region of interest.

The raster version of the RGI is provided to facilitate applications which can consider glaciers as a separate surface type, but can not work with the vector data (shape files) provided so far. Details of the related vector-raster conversion are described in the Algorithm Theoretical Basis Document (ATBD) [RD5]. The raster version of the RGI is provided in netCDF format. The values give the relative glacier cover per grid cell in per cent. This is the most common way of integrating this information, e.g. in a land surface scheme assimilated by climate models.

2. Technical details of the provided product

2.1. Background

The Copernicus Glacier Distribution Service is providing two types of datasets, one is in a vector format, and containing glacier outlines with attribute information (shape files with polygon topology) and a related hypsometry file in csv format and one is in a raster format (netCDF), containing the percentage of glacier coverage per grid cell. The basic structure and contents of the vector file is adopted from the Randolph Glacier Inventory (RGI, see Pfeffer et al. 2014). The technical details presented here are valid for both RGIv5.0 and RGIv6.0 [RD1].

2.2. Technical Specification of the dataset

The RGI is provided as shape files containing the outlines of glaciers in geographic coordinates (longitude and latitude, in degrees), which are referenced to the WGS84 datum. Data are organized by first-order region. For each region there is one shape file (.shp with accompanying .dbf, .prj and .shx files) containing all glaciers and one ancillary .csv file containing all hypsometric data. The attribute (.dbf) and hypsometric files contain one record per glacier. Each object in the RGI conforms to the data-model conventions of ESRI ArcGIS shape files. That is, each object consists of an outline encompassing the glacier, followed immediately by outlines representing all of its nunataks (ice-free areas enclosed by the glacier). In each object successive vertices are ordered such that glacier ice is on the right,i.e. the outline has a clockwise rotation.

2.3. Data fields and hypsometry file

The following attributes are provided with the dataset: GLIMS-ID, RGI-ID, first and second order RGI region, glacier name, area (size in km²), begin and end date, minimum, median and maximum elevation, length, slope, aspect, latitude, longitude, reference, principal investigator (PI), sponsoring agency, and publication. As an additional file (csv format) the area-elevation distribution (hypsometry) is provided for each glacier in 50 or 100 m bins. They are described in the following in more detail. Tables 2 and 3 provide an overview of their characteristics. Please note that this is a selection of attributes. For the full information please download the individual files at glims.org/RGI.

GLIMS-ID

A unique 14-character identifier in the GLIMS format GxxxxxxEyyyyyΘ, where xxxxxx is longitude east of the Greenwich meridian in millidegrees, yyyyy is north or south latitude in millidegrees, and Θ is N or S depending on the hemisphere. The coordinates of GLIMS-ID agree with CenLon and CenLat. Note that, even after the correction of former external GLIMS_IDs described in the next paragraph, GLIMS-IDs in the RGI are provisional. When RGI glaciers are incorporated into GLIMS, an existing GLIMS id code, if there is one, will replace the RGI code.

RGI-ID

A 14-character identifier of the form RGIvv-rr.nnnnn, where vv is the version number, rr is the first-order region number and nnnnn is an arbitrary identifying code that is unique within the region. These codes were assigned as sequential positive integers at the first-order (not second-order) level, but they should not be assumed to be sequential numbers, or even to be numbers. In general the identifying code of each glacier, nnnnn, should not be expected to be the same in different RGI versions. The RGI ID is used as the main identifier of the RGI.

O1Region, O2Region

The codes of the 1^st and 2^nd-order regions of the RGI to which the glacier belongs (GTN-G 2017).

Name

Name of the glacier, or the WGI or WGI-XF id code (modified after Müller et al. 1978) if available. Many glaciers do not have names, and coverage of those that do is incomplete. Of the 211,181 glaciers in the RGI, 39,570 have information in their Name field, although for many the content is actually an id code.

Area

Area of the glacier in km², calculated in cartesian coordinates on a cylindrical equal-area projection of the authalic sphere of the WGS84 ellipsoid, or, for nominal glaciers, accepted from the source inventory.

BgnDate, EndDate

The dates of the source from which the outline was taken, in the form yyyymmdd, with missing dates represented by -9999999. (The form for missing dates was -9990000 in RGI 3.0 and earlier.) When a single date is known, it is assigned to BgnDate. If only a year is given, mmdd is set to 9999. Only when the source provides a range of dates is EndDate not missing, and in this case the two codes together give the date range. In version 5.0, 98% of glaciers (by area; 99% by number) have date information. 85% of the ranges are shorter than four years. Many of the ranges of three years (36-47 months) are from the 1999–2003 period between the launch of Landsat 7 and the failure of the scan-line corrector of its Enhanced Thematic Mapper Plus (ETM+) sensor. In the case several scenes have to be used for an outline, the date of the scene responsible for the majority of the outline will be used or a date range is given.

Zmin, Zmax

Minimum and maximum elevation (m above sea level) of the glacier, obtained in most cases directly from a DEM covering the glacier. For most of the nominal glaciers Zmin and Zmax were taken from the parent inventory, WGI or WGI-XF.

Zmed

Median elevation (m) of the glacier, chosen by sorting the elevations of the DEM cells covering the glacier and recording the 50th percentile of their cumulative frequency distribution. The mean elevation of the glacier is not provided explicitly in the RGI but can be recovered with fair accuracy from the hypsometric list.

Slope

Mean slope of the glacier surface (deg), obtained by averaging single-cell slopes from the DEM.

Aspect

The aspect (orientation) of the glacier surface (deg) is presented as an integer azimuth relative to 0° at due north. The aspect sines and cosines of each of the glacier’s DEM grid cells are summed and the mean aspect is calculated as the arctangent of the quotient of the two sums.

Lmax

Length (m) of the longest surface flowline of the glacier. The length is measured with the algorithm of Machguth and Huss (2014). Briefly, points on the glacier outline at elevations above Zmed are selected as candidate starting points and the flowline emerging from each candidate is propagated by choosing successive DEM cells according to an objectively weighted blend of the criteria of steepest descent and greatest distance from the glacier margin. The latter criterion can be understood as favouring “centrality”, especially on glacier tongues. The longest of the resulting lines is chosen as the glacier’s centreline. In Alaska, Lmax was calculated, only for glaciers larger than 0.1 km², as in Kienholz et al. (2014).

CenLon, CenLat

Longitude and latitude, in degrees, of a single point representing the location of the glacier. These coordinates agree with those in the GLIMS-ID.

Principal investigator (PI)

This field names the person(s) that has(ve) performed the analysis.

Sponsoring agency (Funding)

This fields names the agency that provided the funding for the work

Reference

Here publication(s) related to the work are listed.

Citation

This field is listing how the full dataset has to be cited. Where feasible and provided, regional data extracts should also cite the publication related to the regional study.

2.3.1 Hypsometry

The hypsometry list for each glacier, preceded by copies of the glacier’s RGI-ID, GLIMS-ID and area, is a comma-separated series of elevation-band areas in the form of integer thousandths of the glacier’s total area (see Table 3). The sum of the elevation-band areas is constrained to be 1000. This means that an elevation band’s value divided by 10 represents the elevation band’s area as a percentage of total glacier area. The elevation bands are all 50 m in height and their central elevations are listed in the file header record. Within each hypsometry file the elevation bands extend from 0–50 m up to the highest glacierized elevation band of the first-order region.

The hypsometry for Alaska was provided by Kienholz et al. (2014), relying on the Shuttle Radar Topography Mission DEM (SRTM) south of 60°N. North of 60°N, the elevation sources were a regional interferometric synthetic aperture radar DEM, a DEM from stereographic Satellite Pour l’Observation de la Terre (SPOT) satellite imagery, and the ASTER GDEM2 (version 2 of ASTER GDEM). The hypsometry for the Antarctic and Sub-Antarctic is from Bliss et al. (2013). The primary DEM source was the DEM of the Radarsat Antarctic Mapping Project, with reliance also on the SRTM DEM and ASTER GDEM2, and on maps for some of the Sub-Antarctic islands. Elsewhere the hypsometry was provided by M. Huss, relying on the SRTM DEM between 55°S and 60°N and the ASTER GDEM2 and Greenland Mapping Project (GIMP) DEM north of 60°N (Huss and Farinotti 2012).

Table 2: Overview of the attribute data provided with the area/outlines dataset.

Table 3: Overview of the entries in the hypsometry file.

The information provided with the raster version of the RGI is given in Table 4. The value given is the percentage of glacier cover per 0.1° by 0.1° grid cell.

Table 4: Overview of the attribute data provided with the raster version of the RGI.

2.4. Product examples

Below, we illustrate with selected examples how the vector datasets look and how they can be displayed. Table 5 illustrates a subset of the information stored in the attribute table of the shape file for each glacier, Table 6 reveals how the hypsometric information is stored in the csv file. Apart from just overlaying glacier outlines on a satellite image or creating a mask excluding all regions covered by glaciers (required for uncertainty analysis with stable terrain), one can also colour-code the range of values for a specific attribute within the glacier outlines (Figure 3a) or as filled circles (Figure 3b). The values can also be arranged in histograms, for example to reveal the size-class distribution of glaciers in a region (Figure 4a) or in scatterplots (Figure 4b), revealing possible dependencies among selected attributes. The topographic information can be used to derive several further glacier characteristics (e.g. Haeberli and Hoelzle 1995) or climatic information (Sakai et al. 2015). The hypsometric information (area distribution with elevation) forms the key input when modeling future glacier evolution from climate data, be it with very simplified approaches (e.g. Paul et al. 2007) or more advanced techniques (e.g. Huss and Hock 2015).

Table 5: Screenshot of the information provided in the attribute table of each glacier.

Table 6: The hypsometric information in the csv file is stored per glacier (line) and elevation band (columns).

Figure 3: a) Colour-coded mean glacier aspect for glaciers of the Jostedalsbreen Ice Cap in southern Norway. b) Median elevation of glaciers in the Alps represented as colour-coded circles (from Paul et al. 2020).

Figure 4: a) Histogram showing the size class distribution of glaciers in the Alps. b) Glacier aspect vs. mean elevation for glaciers in the Alps (both panels from Paul et al. 2020).

2.5. Data sources

The glacier outlines in the RGI are derived from a variety of sources by a global community (see Pfeffer et al. 2014). Whereas a main input data source are satellite data (Landsat, ASTER, SPOT), analysts have also digitized outlines from topographic maps, aerial photography and other sources. A major input dataset for the RGI stems from the GLIMS database that has been compiled by a large number of participants since 2000 (e.g. Kargel et al. 2005). Whereas GLIMS is a multi-temporal database containing all outlines being made available, the RGI is a snap-shot in time referring only to one dataset. However, even within a small region glacier outlines can refer to different points in time as clouds may cover parts of a satellite scene and require mosaicking. In part, also the outline of an individual glacier can be composed from several scenes spanning several years. For this reason, a Begin and End date is provided for each glacier in the attribute table. In a few cases, however, the entries in this field are guesses as no information was provided by the analyst.

As a general rule, outlines are derived from satellite scenes acquired around the year 2000 (from Landsat ETM+) to have a good temporal match with the SRTM DEM (that was acquired in February 2000). A common date of all outlines would also be desirable from a modelling perspective, but in reality the outlines in the RGI span a time period of 50 years (1960 to 2010) centred around the years 2000 and 2010.

Glacier outlines that were separated from their neighbours when received were accepted without change, subject only to initial quality control. However, many glacier outlines were originally obtained or contributed as glacier complexes, that is, as collections of contiguous glaciers that meet at glacier divides but not being separated. We used semi-automated algorithms (Bolch et al. 2010, Kienholz et al. 2013) to separate these complexes into glaciers using a watershed algorithm applied to the most appropriate DEM available. The quality of raw output from the algorithms primarily depends on the quality of the DEM used to calculate the divides. Even when a high-quality DEM is available, the algorithm output requires intense manual checking and corrections. These checks were carried out in detail only in a few regions. Elsewhere, in many cases further work is necessary to inspect the quality of drainage divides. We will also use scenes from Sentinel-2A/B to create glacier outlines and the new ArcticDEM, TanDEM-X DEM, and Advanced Land Observing Satellite (ALOS) World 3D - 30m (AW3D30) DEM (as available and appropriate) to derive drainage divides and topographic parameters for each glacier.

2.6. Data Citation requirement

The following reference must be cited when using the RGI version 6.0: RGI Consortium (2017) [RD1].

3.Data access information

The RGI v6.0 provided to the CDS can be found here (https://www.glims.org/RGI/index.html) in its original form, and are described in a related documentation [RD1]. Shape files as well as the raster dataset can be visualised by a range of commercial (e.g. ESRI ArcGIS) and freely available (e.g. QGIS) software packages. An overview on the available datasets (GLIMS) is also possible using the GTN-G browser that can be found at https://www.gtn-g.ch/data_browser. Data will be made available through the Copernicus Climate Data Store (CDS). Registration (free) is required to access the CDS and its toolbox software suite. Data can be downloaded from the website and used under the License ... (included on download page). All requests for information or further data should be channelled through the CDS Knowledge Base.

References

Bliss, A., R. Hock and J.G. Cogley, 2013, A new inventory of mountain glaciers and ice caps for the Antarctic periphery, Annals of Glaciology, 54(63), 191-199.

Bolch, T., B. Menounos, and R. Wheate (2010): Landsat-based glacier inventory of western Canada, 1985–2005. Remote Sensing of Environment, 114, 127-137.

Frey, H. and Paul, F. (2012): On the suitability of the SRTM DEM and ASTER GDEM for the compilation of topographic parameters in glacier inventories. International Journal of Applied Earth Observation and Geoinformation, 18, 480-490.

GCOS (2016): GCOS Implementation Plan 2016. GCOS-200. WMO, Geneva, 315  pp. 

GTN-G 2017: GTN-G Glacier Regions. Global Terrestrial Network for Glaciers. Online: http://dx.doi.org/10.5904/gtng-glacreg-2017-07.

Haeberli, W. and Hoelzle, M. (1995) Application of inventory data for estimating characteristics of and regional climate-change effects on mountain glaciers: A pilot study with the European Alps. Annals of Glaciology, 21, 206–212.

Huss, M., and D. Farinotti, 2012, Distributed ice thickness and volume of all glaciers around the globe, Journal of Geophysical Research, 117, F04010; doi.org/10.1029/2012JF002523.

Huss, M., & Hock, R. (2015). A new model for global glacier change and sea-level rise. Frontiers of Earth Science, 3, 1–22; doi.org/10.3389/feart.2015.00054

IGOS (2007). Integrated Global Observing Strategy Cryosphere Theme Report - For the monitoring of our environment from space and from Earth.

Kargel, J. S., Abrams, M. J., Bishop, M. P., Bush, A., Hamilton, G., Jiskoot, H., ... & Wessels, R. (2005). Multispectral imaging contributions to global land ice measurements from space. Remote Sensing of Environment, 99 (1-2), 187-219

Kienholz, C., R. Hock and A.A. Arendt (2013): A new semi­automatic approach for dividing glacier complexes into individual glaciers. Journal of Glaciology, 59 (217), 925-937.

Kienholz, C., J.L. Rich, A.A. Arendt and R. Hock, 2014, A new method for deriving glacier centerlines applied to glaciers in Alaska and northwest Canada. The Cryosphere, 8, 503–519.

Machguth, H., and M. Huss, 2014, The length of the world’s glaciers – a new approach for the global calculation of center lines, The Cryosphere, 8, 1741–1755.

Müller, F., T. Caflisch and G. Müller, 1978, Instructions for Compilation and Assemblage of Data for a World Glacier Inventory. Supplement: Identification/Glacier Number. Temporary Technical Secretariat for World Glacier Inventory, Department of Geography, Federal Institute of Technology, Zürich.

Paul, F., Kääb, A., Maisch, M., Kellenberger, T. W. and Haeberli, W. (2002): The new remote-sensing-derived Swiss glacier inventory: I. Methods. Annals of Glaciology, 34, 355-361.

Paul, F., Maisch, M., Rothenbühler, C., Hoelzle, M., & Haeberli, W. (2007). Calculation and visualisation of future glacier extent in the Swiss Alps by means of hypsographic modelling. Global and Planetary Change, 55(4), 343–357.

Paul, F., R. Barry, J.G. Cogley, H. Frey, W. Haeberli, A. Ohmura, S. Ommanney, B.Raup, A. Rivera, M. Zemp (2009): Recommendations for the compilation of glacier inventory data from digital sources. Annals of Glaciology, 50 (53), 119-126.

Paul, F., N. Barrand, E. Berthier, T. Bolch, K. Casey, H. Frey, S.P. Joshi, V. Konovalov, R. Le Bris, N. Mölg, G. Nosenko, C. Nuth, A. Pope, A. Racoviteanu, P. Rastner, B. Raup, K. Scharrer, S. Steffen and S. Winsvold (2013): On the accuracy of glacier outlines derived from remote sensing data. Annals of Glaciology, 54 (63), 171-182.

Paul, F., S.H. Winsvold, A. Kääb, T. Nagler and G. Schwaizer (2016): Glacier Remote Sensing Using Sentinel-2. Part II: Mapping Glacier Extents and Surface Facies, and Comparison to Landsat 8. Remote Sensing , 8(7), 575; doi:10.3390/rs8070575.

Paul, F., Rastner, P., Azzoni, R.S., Diolaiuti, G., Fugazza, D., Le Bris, R., Nemec, J., Rabatel, A., Ramusovic, M., Schwaizer, G., and Smiraglia, C. (2020): Glacier shrinkage in the Alps continues unabated as revealed by a new glacier inventory from Sentinel-2. Earth Systems Science Data, 12(3), 1805-1821.

Pfeffer, W.T., 18 others and the Randolph Consortium (2014): The Randolph Glacier Inventory: a globally complete inventory of glaciers. Journal of Glaciology, 60 (221), 537-552.

RGI Consortium (2017): Randolph Glacier Inventory – A Dataset of Global Glacier Outlines: Version 6.0, GLIMS Technical Report, 71 pp. Online: https://glims.org/RGI/00_rgi60_TechnicalNote.pdf (last access: 21 Nov 2022).

Sakai, A., Nuimura, T., Fujita, K., Takenaka, S., Nagai, H., and Lamsal, D. (2015): Climate regime of Asian glaciers revealed by GAMDAM glacier inventory. The Cryosphere, 9, 865–880.

Vaughan, D.G., J. C. Comiso, I. Allison, J. Carrasco, G. Kaser, R. Kwok, P. Mote, T. Murray, F. Paul, J. Ren, E. Rignot, O. Solomina, K. Steffen and T. Zhang (2013): Observations: Cryosphere. In: Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the IPCC. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, pp. 317-382.

Zemp, M., Thibert, E., Huss, M., Stumm, D., Rolstad Denby, C., Nuth, C., Nussbaumer, S.U., Moholdt, G., Mercer, A., Mayer, C., Joerg, P.C., Jansson, P., Hynek, B., Fischer, A., Escher-Vetter, H., Elvehøy, H., and Andreassen, L.M. (2013): Reanalysing glacier mass balance measurement series. The Cryosphere, 7, 1227-1245.

Related articles

• Page:
  ECV - Glaciers, Ice Sheets, Lakes, Soil Moisture
• Page:
  IV Version 1.2; SEC Version 2.0; GMB Version 2.0: Product User Guide Specification (PUGS)
• Page:
  IV Version 1.2; SEC Version 2.0; GMB Version 2.0: System Quality Assurance Document (SQAD)
• Page:
  IV Version 1.2; SEC Version 2.0; GMB Version 2.0: Product Quality Assurance Document (PQAD)
• Page:
  IV Version 1.2; SEC Version 2.0; GMB Version 2.0: Product Quality Assessment Report (PQAR)