Extending glacier monitoring into the Little Ice Age and beyond

: Reconstructions of glacier front variations based on well-dated historical evidence from the Alps, Scandinavia, and the southern Andes, extend the observational record as far back as the 16th century. The standardized compilation of paleo-glacier length changes is now an integral part of the internationally coordinated glacier monitoring system. Reconstructions of glacier front variations based on well-dated historical evidence from the Alps, Scandinavia, and the southern Andes, extend the observational record as far back as the 16 th century. The standardized compilation of paleo-glacier length changes is now an integral part of the internationally coordinated glacier monitoring system.


Implications for interpretation of C/N, δ 13 C and δ 15 N values
Bias by acid pre-treatment on OM can signi cantly undermine C/N values as indicators of OM provenance. For example, Figures 1 and 2 show that although broccoli (Brassica oleracea) is a terrestrial C 3 plant, an aquatic or aquatic/terrestrial combination could be concluded from the data, depending upon the method and/or acid reagent (see Brodie et al., 2011a). In addition, C/N values can also vary considerably depending on whether they are calculated with %N from treated or untreated sample aliquots (see Brodie et al., 2011a, b). For δ 13 C, biasing in the range of 0.2 -7.1‰ can undermine C 3 vs. C 4 plant type interpretations, and together with C/N undermine bi-plot interpretations of C/N, δ 13 C and δ 15 N values. This clearly demonstrates that the data are inherently unreliable as a function of the analytical approach. Although the underlying mechanisms require further research, it is clear the biases represented here across a range of terrestrial and aquatic, modern and ancient or-ganic materials has direct implications for paleo reconstructions: understanding and reducing the uncertainty on the data is an essential prerequisite for reliable interpretations and reconstructions.

Concluding Remarks
The systematic comparisons of Brodie et al. (2011a, b) clearly demonstrate nonlinear and unpredictable biasing of OM due to acid pre-treatment, and concomitantly indicate that complete IC removal (the purpose of acid pre-treatment) is not guaranteed. It is concluded that these biases are inherently not correctable but inevitable, and have a direct consequence for the accuracy and precision of measured values (i.e., signi cantly greater than instrument precision). Moreover, environmental interpretations of the data in both modern and paleo systems could be highly questionable.

References
Brodie, C.R., Leng, M.J., Casford, J.S.L., Kendrick, C.P., Lloyd, J.M., Zong, Y.Q. and Bird, M.I., 2011a: Evidence for bias in C and N concentrations and d 13 C composition of terrestrial and aquatic organic materials due to pre-analysis acid preparation methods, Reconstructions of glacier front variations based on well-dated historical evidence from the Alps, Scandinavia, and the southern Andes, extend the observational record as far back as the 16 th century. The standardized compilation of paleo-glacier length changes is now an integral part of the internationally coordinated glacier monitoring system.
Glaciers are sensitive indicators of climatic changes and, as such, key targets within the international Global Climate Observing System (GCOS, 2010). Glacier dynamics contribute signi cantly to global sea level variations, alter the regional hydrology, and determine the vulnerability to local natural hazards. The worldwide monitoring of glacier distribution and uctuations has been well established for more than a century (World Glacier Monitoring Service, 2008). Direct measurements of seasonal and annual glacier mass balance, which are available for the past six decades, allow us to quantify the response of a glacier to climatic changes. The variations of a glacier front position represents an indirect, delayed, ltered and enhanced response to changes in climate over glacierspeci c response times of up to several decades ( Zumbühl, 1980;Zumbühl et al., 1983;Karlén, 1988;Zumbühl and Holzhauser, 1988;Luckman, 1993;Tribolet, 1998;Nicolussi and Patzelt, 2000;Holzhauser et al., 2005;Nussbaumer et al., 2007;Zumbühl et al., 2008;Masiokas et al., 2009;Nesje, 2009;Holzhauser, 2010;Nussbaumer and Zumbühl, 2011). However, the majority of the data remains inaccessible to the scienti c community, which limits the veri cation and direct comparison of the results. In this article, we document our rst attempt towards standardizing reconstructed glacier front variations and integrating them with in situ measurement data of the World Glacier Monitoring Service (WGMS).

Standardization and database
The standardization of glacier front variations is designed to allow seamless comparison between reconstructions and in situ observations while still providing the most relevant information on methods and uncertainties of the individual data series. The standardized compilation of in situ observations is straightforward: the change in glacier front position is determined between two points in time and supplemented by information on survey dates, methods and data accuracies. The reconstruction of paleo-glacier front positions and their dating is usually more complex and based on multiple sources of evidence. Spatial uncertainty can arise from ambiguous identi cation of the glacier terminus, while temporal uncertainty is associated with the dating technique used in each case. A good example is the use of oil paintings, where it is important to distinguish between the time of rst draft (e.g., a landscape drawing in the eld) and production of the painting itself, which could happen several years later.
The concept of integration of reconstructed front variations into the relational glacier database of the WGMS was jointly developed by natural and historical scientists. The glacier reconstruction data are stored in two data tables. The rst table contains summary information of the entire reconstruction series including a gure of the cumulative length changes, investigator information, and references. The second table stores the individual glacier front variation data, minimum and maximum glacier elevation, and metadata related to the reconstruction methods and uncertainties. The metadata are stored in a combination of prede ned choices of methods and open text elds. This ensures a standardized description of the data, but also provides space for individual remarks. Both data tables are linked by a unique numeric glacier identi er to the main table of the WGMS database with general information of the present-day glacier. This database relationship enables the direct comparison of reconstructed with observed front variations. Figure 1 shows an example for the Mer de Glace glacier with the sum of available meta-information stored within the database.

Results and discussion
The WGMS database contains detailed inventory information of about 100,000 glaciers worldwide. In addition, in situ observations of frontal variations of 1,800 glaciers, mass balance data of 230 glaciers, and reconstructed frontal variations of the 26 glaciers mentioned below are now readily available in standardized and digital form. The reconstructed front variations extend the direct observations (mostly from the 20 th century) by two centuries in Norway and by four centuries in the Alps and South America. Also available are moraines data back to the mid-Holocene. The direct comparison of long-term reconstructions with in situ observations reveals some striking features (Fig. 2). The investigated glaciers in the western and central Alps show several periods of marked advances during the LIA, that are similar or even larger than LIA extent around AD 1850. Reconstructions reveal dramatic glacier advances that started in the late 16 th century, overran cropland and hamlets, and reached maximum lengths at ca. AD 1600 and 1640 (Zumbühl, 1980;Nussbaumer et al., 2007;Steiner et al., 2008). Further maxima in glacier extent were reached around AD 1720, 1780, 1820 and 1850. After this last event, the direct measurement series show an impressive glacier retreat of 1-3 km, with intermittent minor re-advances in the 1890s, 1920s and between AD 1965 and 1985.
The historically based reconstructions for glaciers in southern Norway (Nussbaumer et al., 2011;Nesje, 2009) indicate that the maximum glacial extent of the LIA peaked around AD 1750 at Jostedalsbreen and in the late 1870s at Folgefonna. The reconstructed front variation series of Nigardsbreen (an outlet glacier of Jostedalsbreen) shows that the maximum LIA expansion occurred in AD 1748, and that it was preceded by a period of strong frontal advance that lasted (at least) 3-4 decades. Other sources report devastation of pastures by neighboring glaciers at that time (Nussbaumer et al., 2011). Direct observations at Nigardsbreen and other glaciers in southern Norway reveal a regional pattern of recent intermittent re-advances mainly during the 1990s. It is worthwhile to note that the extent of Nigardsbreen in the 17 th century was similar to that of the present day (Nesje et al., 2008), and that the period of glacier re-advances in the 1990s is short (both in time and extension) considering the overall retreat of 1-4 km since the LIA maximum.
In southern South America, datable evidence for past glacier variations is most abundant in the Patagonian Andes. Despite recent e orts in this region and in other Andean sites to the north (Espizua, 2005;Espizua and Pitte, 2009;Jomelli et al., 2009;Masiokas et al., 2009), the number of chronologies of glacier uctuations prior to the 20 th century is still limited. In the southern Andes, most records indicate that glaciers were generally more extensive prior to the 20 th century, with dates of maximum expansion ranging from the 16 th to the 19 th centuries. Based on the available evidence, Glaciar Frías (northern Patagonian Andes) probably contains the best-documented history of uctuations since the LIA in southern South America, covering the period between AD 1639-2009.

Conclusions and outlook
Reconstructed LIA and Holocene glacier front variations are crucial for understanding past glacier dynamics and enable an objective assessment of the relative signi cance of recent glacier uctuations (i.e., 20 th and late 19 th century changes) in a long-term context. The standardized compilation and free dissemination of reconstructed and in situ observed glacier uctuation records o er several bene ts for both data providers and users. Their incorporation within the international glacier databases guarantees the long-term availability of the data series and increases the visibility of the scienti c results (which in historical glaciology are often the work of a lifetime). Furthermore, the database facilitates comparisons between glaciers and between di erent methods, and opens the eld to numerous scienti c studies and applications.
As the next steps of this new initiative, we aim to: (1) integrate a greater number of time series, (2) incorporate records that cover the entire Holocene, and (3) include data from other regions (e.g., the Himalayas, North America). Ideally, the growing new dataset will facilitate collaboration between the glacier monitoring and reconstruction communities and become an additional tool for the comparison of present-day to pre-industrial climate changes. This should eventually result in new scienti c ndings (e.g., related to the climatic interpretation of past and present glacier uctuations).

International glacier databases: Requesting and submitting data
Worldwide collection of standardized data on the distribution and changes of glaciers has been internationally coordinated since 1894. Today, the World Glacier Monitoring Service (www.wgms.ch) is in charge of the compilation and dissemination of glacier datasets in close collaboration with the US National Snow and Ice Data Center (www.nsidc.org) and the Global Land Ice Measurement from Space initiative (www.glims.org) within the framework of the Global Terrestrial Network for Glaciers (www.gtn-g.org).
For available data or guidelines on data submission please check these websites and/or contact us directly: wgms@ geo.uzh.ch