Following recommendations from the International Commission on Snow and Ice for a world glacier inventory, an inventory of glaciers in China was carried out by Chinese glaciologists from 1978 to 2002. Each glacier was measured from aerial photographs and topographical maps and 34 parameters recorded. These parameters were then analyzed statistically for the various river systems in China. Twelve volumes of the Glacier Inventory of China (GIC) have been published, consisting of 22 parts in 21 books. The data were subsequently abridged into a Concise GIC, published in Chinese (2005) and in English (2008), to make the glacier inventory more accessible and better adapted for assessing glacier response to climate change. After the GIC was completed, new aerial photographs became available and remote-sensing techniques became more common. To investigate glacier changes since completion of the first GIC, a second Glacier Inventory of China was initiated in 2007. This 5 year project, supported by the Ministry of Science and Technology, will be undertaken mainly using remote-sensing techniques.

This paper reviews the origins of the World Glacier Inventory program and Canada’s participation. It tracks the program of the Canadian federal government from its initiation to its demise. Today, progress is geared towards the international GLIMS and GlobGlacier programs, to which scientists from several Canadian universities are making contributions. Achievements during the past decade, towards providing data on Canada’s glacierized areas in the Arctic and in western Canada, are documented.

The quantitative assessment of glacier changes as well as improved modeling of climate-change impacts on glaciers requires digital vector outlines of individual glacier entities. Unfortunately, such a glacier inventory is still lacking in many remote but extensively glacierized gions such as the Canadian Arctic. Multispectral satellite data in combination with digital elevation models (DEMs) a particularly useful for creating detailed glacier inventory data including topographic information for each entity. In this study, we extracted glacier outlines and a DEM using two adjacent Terra ASTER scenes acquired in August 2000 for a remote region on southern Baffin Island, Canada. Additionally, Little Ice Age (LIA) extents we digitized from trimlines and moraines visible on the ASTER scenes, and Landsat MSS and TM scenes from the years 1975 and 1990 we used to assess changes in glacier length and area. Because automated delineation of glaciers is based on a band in the shortwave infrared, we have developed a new semi-automated glacier-mapping approach for the MSS sensor. Wrongly classified debris-coved glaciers, water bodies and attached snowfields we corrected manually for both ASTER and MSS. Glacier drainage divides we manually digitized by combining visual interptation with DEM information. In this first paper, we describe the applied methods for glacier mapping and the glaciological challenges encounted (e.g. data voids, snow cover, ice caps, tributaries), while the second paper ports the data analyses and the derived changes.

Despite its large area covered by glaciers and ice caps, detailed glacier inventory data are not yet available for most parts of Baffin Island, Canada. Automated classification of satellite data could help to overcome the data gaps. Along-track stereo sensors allow the derivation of a digital elevation model (DEM) and glacier outlines from the same point in time, and are particularly useful for this task. While part I of this study describes the remote-sensing methods, in part II we present an analysis of the derived glacier inventory data for 662 glaciers and an application to glacier volume and volume-change calculations. Among other things, the analysis reveals a mean glacier elevation of 990 m, with a weak dependence on aspect and a close agreement of the arithmetic mean with the statistical mean elevation as derived from the DEM. A strong scatter of mean slope is observed for glaciers <1 km2, and the derived glacier thickness differs by a factor of two for glaciers of the same size. For the period from about 1920 to 2000 the relative area change is –12.5% (264 glaciers), with a strong dependence on glacier size. Mean mass loss as derived from volume changes is about –0.15 mw.e. a–1.

The World Glacier Inventory (WGI) was conceived half a century ago as an activity to be completed during the International Geophysical Year, 1957/58. It consisted until very recently of nearly 70 000 glacier records covering slightly less than one-quarter of the glacier ice outside the ice sheets. A complete WGI must be a compromise if it is to be available and usable soon. A more complete version, called WGI-XF, is available and usable now and contains records for just over 131 000 glaciers and nearly half of the global extent of ice. The additional glaciers come mainly from the assimilation of existing inventories but also from rescuing inventories that have been lost and from new inventories in Canada and the Subantarctic. In WGI-XF, the XF stands for ‘extended format’, flagging the fact that WGI-XF conforms to a set of explicit specifications which enhance usefulness by eliminating low-level inconsistencies. Two important features are nominal glaciers and glacier complexes. A nominal glacier, of which there are about 5000 in WGI-XF, is one about which little is known other than its existence and approximate location. A glacier complex is one or more contiguous glaciers. This term embodies the idea, which is not new, that inventories can be preliminary, based upon vector outlines which await subdivision by trained glaciologists. Many regional studies have found that measurements of changes in single glaciers require accurate work and painstaking quality control. WGI-XF is not assuredly reliable as a source for such detailed work, but there are several other subjects in which less detail would be a price worth paying for more complete coverage. Incomplete information about the dates of imagery and maps is a hindrance to analysis, and the recovery of dates from metadata should have high priority.

Glacier changes in the Karlik Shan, eastern Tien Shan, from 1971/72 to 2001/02 were monitored in this study. Topographic maps of 1 : 50 000 scale based on aerial photographs from 1971/72 and satellite images (Landsat TM, Landsat ETM+ and ASTER) from 1992, 2001 and 2002 were used to map glacier extent through a process of manual digitizing. The total glacier area decreased by 5.3% from 1971/72 to 2001/02. The rate of glacier area shrinkage was 0.13% a–1 between 1972 and 1992, but it was 0.27% a–1 from 1992 to 2001/02, suggesting accelerated glacier retreat in recent decades. Glacier changes in the region are a response to summer temperature increase. Annual precipitation also showed an upward trend, but this could not compensate for the mass loss due to ablation.

A new approach to glacier inventory, based on airborne laser-scanner data, has been applied to South Tyrol, Italy: it yields highly accurate results with a minimum of human supervision. Earlier inventories, from 1983 and 1997, are used to compare changes in area, volume and equilibrium-line altitude. A reduction of 32% was observed in glacier area from 1983 to 2006. Volume change, derived from the 1997 and 2006 digital elevation models, was –1.037 km3, and an ELA rise of 54 m, to almost 3000 m a.s.l., was calculated for this period. Losses vary widely for individual glaciers, but have accelerated for all South Tyrolean glaciers since the first inventory in 1983.

On 16–18 June 2008 the US National Snow and Ice Data Center held a GLIMS workshop in Boulder, CO, USA, focusing on formulating procedures and best practices for operational glacier mapping using satellite imagery. Despite the progress made in recent years, there still remain many cases where automatic delineation of glacier boundaries in satellite imagery is difficult, error prone or time-consuming. This workshop identified six themes for consideration by focus groups: (1) mapping clean ice and lakes; (2) mapping ice divides; (3) mapping debris-covered glaciers; (4) assessing changes in glacier area and elevation through comparisons with older data; (5) digital elevation model (DEM) generation from satellite stereo pairs; and (6) accuracy and error analysis. Talks presented examples and work in progress for each of these topics, and focus groups worked on compiling a summary of available algorithms and procedures to address and avoid identified hurdles. Special emphasis was given to establishing standard protocols for glacier delineation and analysis, creating illustrated tutorials and providing source code for available methods. This paper summarizes the major results of the 2008 GLIMS workshop, with an emphasis on definitions, methods and recommendations for satellite data processing. While the list of proposed methods and recommendations is not comprehensive and is still a work in progress, our goal here is to provide a starting point for the GLIMS regional centers as well as for the wider glaciological community in terms of documentation on possible pitfalls along with potential solutions.

Automated glacier mapping from thresholded band ratios of multispectral satellite data is a well-established technique to update glacier inventories over large and remote regions. The local glaciers and ice caps on Greenland are of particular interest for such efforts, as they have been only partly mapped, mainly during the 1940s–60s, and their potential contribution to global sea-level rise could be large. Here we use three Landsat ETM+ scenes from 2001 covering Disko Island (Qeqertarsuaq) and the Nuussuaq and Svartenhuk peninsulas, West Greenland, to map the glacier extent in 2001 of 1172 entities. We also manually digitize Little Ice Age (LIA) extents from clearly visible trimlines for a subsample of 500 entities. In this region with numerous surge-type glaciers, the related area-change calculation is challenging and we consider different samples with and without known surging glaciers. For the three regions the mean area changes are –28%, –20% and –23%, respectively, when known surge-type glaciers are excluded. The glaciers on smaller islands and peninsulas closer to the margin of the ice sheet show a lower mean area change of –15%. Moreover, lower (–16%) and upper (–21%) bounds are calculated for the overall area changes in the entire region between the LIA and 2001 using different upscaling assumptions. Cumulative length changes since the LIA are found to be slightly lower for surge-type glaciers.

Recent climate changes have had a significant impact on the high-mountain glacial environment. Rapid melting of glaciers has resulted in the formation and expansion of moraine-dammed lakes, creating a potential danger from glacial lake outburst floods (GLOFs). Most lakes have formed during the second half of the 20th century. Glaciers in the Mount Everest (Sagamartha) region, Nepal, are retreating at an average rate of 10–59 ma–1. From 1976 to 2000, Lumding and Imja Glaciers retreated 42 and 34 ma–1, respectively, a rate that increased to 74 ma–1 for both glaciers from 2000 to 2007. During the past decade, Himalayan glaciers have generally been shrinking and retreating faster while moraine-dammed lakes have been proliferating. Although the number of lakes above 3500 m a.s.l. has decreased, the overall area of moraine-dammed lakes is increasing. Understanding the behaviour of glaciers and glacial lakes is a vital aspect of GLOF disaster management.

The ice cover of the Nianchu river basin, southern Tibetan Plateau, was mapped for 2005, using a SPOT5 scene, and for 1990 and 2000 from Landsat TM/ETM. Digital elevation models (DEMs) were generated from 1 : 50 000 scale topographical maps. The results show that in 2005 there were 136 glaciers in this basin, with a total area of 224 km2. Of these, 37 glaciers had an area >1 km2 and 10 were larger than 5 km2; the average snout altitude was 5608m a.s.l. A comparison of outlines from the last 15 years shows that most glaciers have decreased in size; none have advanced. From 1990 to 2005, Xiaquepu glacier No. 56 and Shimozongpu glacier No. 38 retreated 310 and 560 m, respectively. The mean reduction in glacier surface area was 5% (10 a)–1 while the area of glacial lakes expanded by 10%; nine new lakes formed in this basin over the 15 year period. Because air temperatures here have been increasing, while precipitation has remained steady, glacier retreat is considered to be related to rising temperature.

A new glacier inventory of Iran, compiled according to GLIMS guidelines through the use of photogrammetry and remote sensing supported by fieldwork, provides the first comprehensive study of its mountain glaciers. The glaciers are found in five main areas: two in the higher elevations of the Alborz mountain range (Damavand and Takhte–Soleiman regions), two on the Zardkuh and Oshtorankuh mountain chain in the Zagros mountain range and one in the Sabalan Mountains in northwest Iran. Several important glacier attributes, including minimum and maximum height of ice, area and maximum length and width, together with glacier extent, were successfully extracted using aerial and satellite imagery. Thereafter a comprehensive glacier database was established in a GIS environment.

According to the division into subareas of water-resource distribution in China, and based on the Glacier Inventory of China (GIC), China’s total glacier water storage is 5040.2 x 109m3, 33.0% of which is distributed in the southwest drainage basins and 64.1% in the northwest inland drainage basins, forming enormous solid reservoirs with plentiful freshwater storage. Glacier change in China is estimated for the periods from the Little Ice Age (LIA) maximum to about 1960 and from 1960 to 1995. The relative glacierized area loss is 23% and 8.9% respectively for maritime glaciers and 15% and 4.9% respectively for continental glaciers. The normal annual glacier meltwater runoff is estimated at 60 465 x 106m3 by the climate parameter temperature-index method, 38.7% of which is distributed in the northwest inland drainage basins, and at 61 574 x 106m3 by the glacier system temperature-index method, 41.5% of which is distributed in the northwest inland drainage basins. Simulation of glacier meltwater runoff under the temperature change ratio of 0.03 K a–1 by the glacier system model in west China between 1980 and 2000 indicates that the total glacier meltwater runoff increment is 10.8% overall, 14.3% in the inland drainage basins of northwest China and 9.0% in the outflow drainage basins of southwest China.

Results of a new glacier inventory of the upper Huasco valley, which lies within the arid Norte Chico zone of the Chilean Andes, are presented for 2004. Despite the high altitude, the glaciation in this region is limited in extent and is not classical mountain glaciation, which poses difficulties in completing standard inventory attribute tables. Small cornice-style ridgeline features constitute a large number of the non-transient ice bodies identified, and glaciers with surface areas <0.1 km2 comprise 18% of the glacierized area and 3% of the water resource stored as glacier ice within the Huasco valley. Rock glaciers are an important component of the cryosphere, comprising 12% of the total water volume stored in glacial features. Changes in glacier area over the last ~50 years are in line with those for glaciers in central Chile despite the contrasting climate conditions. Projections of glacier area change based on glacier hypsometry and zero isotherm shifts predicted using the PRECIS regional model temperature change for IPCC scenario B2 conditions suggest that the survival of 65% of glacier area and 77% of active rock-glacier area will be threatened under forecast conditions for the end of the 21st century.

Modern geoinformatic techniques allow the automated creation of detailed glacier inventory data from glacier outlines and digital terrain models (DTMs). Once glacier entities are defined and an appropriate DTM is available, several methods exist to derive the inventory data (e.g. minimum, maximum and mean elevation; mean slope and aspect) for each glacier from digital intersection of both datasets. Compared to the former manual methods, the new grid-based statistical calculations are very fast and reproducible. The major aim of this contribution is to help in standardizing the related calculations to enhance the integrity of the Global Land Ice Monitoring from Space (GLIMS) database. The recommendations were prepared by a working group and also contribute to the European Space Agency project GlobGlacier. The document follows the former UNESCO manual for the production of the World Glacier Inventory published in 1970, identifies the potential pitfalls, and describes the differences from the former methods of compilation. The online background material for this paper (see http://www.glims.org) contains example scripts for calculation of each parameter and will be updated when required.

A glacier’s most fundamental altitude is the equilibrium-line altitude (ELA) because it divides the glacier into ablation and accumulation areas. The best parameterization of the ELA for glacier inventory is the balanced-budget ELA. We discuss direct estimation of balanced-budget ELA from mass-balance data for individual glaciers, and indirect estimation of balanced-budget ELA from simple topographic parameters available from the World Glacier Inventory (WGI), i.e. the area-median and maximum and minimum altitudes. Mass balance and ELA for individual glaciers are usually strongly correlated and we calculate balanced-budget ELA from the regression equation linking the two. We then compare balanced-budget ELA with area-median and mid-range altitudes for the 94 glaciers for which we have all the necessary data. The different ELA estimates agree well enough (±82 to ±125 m) to describe geographical variations in ELA and for application of glacier–climate models to glacier inventory data. Mid-range and area-median altitudes are already available for tens of thousands of glaciers in the current WGI and should be evaluated in future inventories.

We compiled a detailed glacier inventory of 176 glaciers in the Clemenceau Icefield Group (CIG) and adjacent Chaba Group (CH), Canada, based on 2001 Landsat 7 and 2000–03 ASTER satellite imagery and Natural Resources Canada digital elevation models. We used this inventory to measure length and mass-balance changes and their possible controls. A classification of glacier hypsometry in the form of a hypsometric index was used to assess the sensitivity of different glacier systems to a unit rise in snowline. The altitude and AAR of possible steady-state ELAs was derived using several methods, and was compared to late-summer snowlines of 2001 .We further compared planar glacier area to slope-corrected area, and compared the effects on the shape of the hypsometric curves, on the total glacier area and on the aspect–area distribution. In 2001, CIG had a glaciated area of 271 km2 and had lost 42 km2 since the mid-1980s. CH had a total area of 69 km2 and had lost 28 km2. Average retreat rates are 14 ma–1 for the period 1850–2001 (n=39) and 21 ma–1 for 1986/87–2001 (n=23), indicating accelerated retreat. Larger glaciers and those that experience tributary detachment tend to retreat faster. The difference between planar and slope-corrected glacier areas ranges from 5% to 20%, with a 6% increase for the entire CIG/CH region. The area increase does not change the shape of the hypsometric curves.

An inventory of the surface area and volume of the world’s glaciers, outside Greenland and Antarctica, was part of the International Hydrological Decade (1965–74). It was considered essential to an understanding of the role played by glaciers in the hydrological cycle and was to be repeated every 50 years to detect change. To date, 46% of the estimated total glacier area has been inventoried and made available through the World Glacier Monitoring Service and the US National Snow and Ice Data Center. As the original inventory method was too time-consuming and inapplicable for some areas, a simplified method was developed in the early 1980s using satellite images. The Global Land Ice Measurements from Space (GLIMS) project now covers 34% of the estimated glacierized area outside Greenland and Antarctica. Both inventory efforts have made good progress and contributed substantially to our knowledge of glaciology and its related sciences, but global coverage is still incomplete. If both inventories are combined, 46% of the world’s glacierized area is still missing; 26% is covered by both methods, which allows the quality of the satellite-based and semi-automatic inventories to be assessed by comparison. About 95 000 glaciers remain to be inventoried, of which about half are in the Canadian Cordillera, South America and the Canadian Arctic Islands. As the cryosphere is changing rapidly, it is of the utmost importance to complete the global glacier inventory as soon as possible, and identify an appropriate repeat cycle.