Elsevier

Global and Planetary Change

Volume 133, October 2015, Pages 109-119
Global and Planetary Change

Mass loss and imbalance of glaciers along the Andes Cordillera to the sub-Antarctic islands

https://doi.org/10.1016/j.gloplacha.2015.08.009Get rights and content

Highlights

  • The mean observed mass-balance became more negative from 1993 through 2012.

  • Glaciers in the northern part of Andes are mostly out of balance with the present climate.

  • Glacier mass-balance on the sub-Antarctic islands distinguishes from that of glaciers in the Andes.

  • South America should be divided into three glacier regions, and not two as earlier stated.

  • The cycle of mass-balance is linked to the PDO and ENSO with an ~ 8 year and 0 year variability.

Abstract

Here, we examine available glacier mass-balance records between 1993 and 2012 for Andes Cordillera, South America (6.5°N–45.8°S), and the sub-Antarctic islands around the northern tip of the Antarctic Peninsula (62.7°S–63.8°S) to determine their recent mass loss and imbalance with the present climate. The mean annual observed mass-balance Ba changed from − 620 ± 390 (1993–2002) to − 740 ± 240 kg m 2 yr 1 (2003–2012) and for this past decade showed a decrease in Ba from south to north. These glaciers had a mean accumulation area ratio of 0.42, which is below the AAR value for glaciers in equilibrium, reflecting mean area and volume imbalances of 23% and 27%, respectively. Glaciers in the northern part of Andes Cordillera are most out of balance with the present climate (33%), while glaciers on the sub-Antarctic islands are only slightly out of balance (4%). We identified a spatiotemporal cycle of Ba that distinguishes glaciers on the sub-Antarctic islands from glaciers in the Andes using an Empirical Orthogonal Function analysis. This analysis also revealed that South America should be divided into three individual glacier regions, and not two regions as earlier stated. Overall, the spatiotemporal cycles identified correlate to the multivariate El Niño Southern Oscillation Index instantaneously (zero-year lag-time) and to the Pacific Decadal Oscillation with an approximately eight-year lag-time.

Introduction

Glaciers are sensitive to changes in climate, especially surface air temperature and precipitation (AMAP, 2011, Chapter 7). Over a year, when the sum of accumulation (mainly due to snowfall) equals the sum of ablation (mainly due to surface melting, runoff, and calving), a glacier is considered to be in annual balance. If the annual accumulation sum is unequal to the ablation sum over a period of years to decades, glaciers are likely to either thin and retreat or thicken and advance. For the last decades most of the Earth's glaciers, including South American glaciers (e.g., Georges, 2004, Casassa et al., 2007, Masiokas et al., 2008, Rabatel et al., 2011, Rabatel et al., 2013), have undergone general thinning and recession (e.g., World Glacier Monitoring Service (WGMS), 2012, IPCC, 2013a, Chapter 4; Leclercq et al., 2014). Glacier annual mass-balance (Ba) observations from the last decades have shown an overall increase in mass loss (Cogley, 2009, Cogley, 2012, Dyurgerov, 2010), resulting in more negative annual Ba during the first pentad (five years) of the first decade of the new millennium (WGMS, 2013). Although more moderate, these negative Ba appear to have been sustained for the second pentad (e.g., Cogley, 2012, Marzeion et al., 2014).

Globally, direct Ba field measurements exist for 340 glaciers, of which 25 are located in South America and on islands around the northern tip of the Antarctic Peninsula. Despite only representing a minor fraction of the real number of glaciers in South America (~ 21,800 glaciers; Radić et al., 2013) and on the sub-Antarctic islands (~ 1000 glaciers; Bliss et al., 2013), this sample of 25 represents an important reference dataset in regard to the health of glaciers in these regions as a whole.

Given the relatively small sample of available direct Ba observations, research has to rely on modeling, remote sensing-derived measurements, and upscaling approaches to define and explore trends in Ba and associated dynamic imbalance conditions. Specifically, these approaches have been used to estimate global mean glacier Ba changes, including Ba changes for South America and the Antarctic Peninsula region (e.g., Kaser et al., 2006, Hock et al., 2009, Gardner et al., 2013, Marzeion et al., 2012, Mernild et al., 2013). Gardner et al. (2013), for example, used satellite gravimetry, altimetry, and direct glaciological records to estimate mean glacier Ba conditions and ice mass contribution to sea-level rise for the northern and southern Andes (2003–2009). They estimated glacier mass losses of − 1080 ± 360 for northern Andes and − 990 ± 360 kg m 2 yr 1 for central and southern Andes (2003–2009). Additionally, for Antarctica and the sub-Antarctic region, Gardner et al. (2013) estimated a mean glacier mass loss of − 50 ± 70 kg m 2 yr 1 (no separate mass budget rates were estimated for the Antarctic Peninsula alone). Specifically for the northern Antarctic Peninsula (< 66°S), a mean mass loss of − 72 kg m 2 yr 1 has been estimated (pers. com. T. Scambos, July 2014; Scambos et al., 2014), based on satellite laser altimetry and satellite stereo-image elevation changes between 2001 and 2010. Based on available direct glaciological Ba observations, Mernild et al. (2013; using n = 8 glaciers) reported a mean glacier mass loss of − 830 ± 280 and − 210 ± 230 kg m 2 yr 1 for the Northern and the Southern Andes, respectively and − 130 ± 100 kg m 2 yr 1 for glaciers peripheral to the Antarctic Peninsula (2001–2010). These estimates take into account uncertainties related to under-sampling errors and geographical biases.

These three aforementioned studies are examples of Ba estimates made for similar time periods but based on different approaches — together they represent a prediction that mass loss is much less in the Antarctic region and much greater in the north of the Andes Cordillera. However, much of this inference is based on interpolation models forced by limited ground-based observation.

On an individual glacier scale, Rignot et al. (2003), Davies and Glasser (2012), Willis et al., 2012a, Willis et al., 2012b, and Schaefer et al., 2013, Schaefer et al., 2015, for example, performed area and mass-balance analysis studies specifically for the Patagonia ice fields using radar observations, satellite-based observations, and simulations, respectively, while Kaser et al. (2003), Francou et al. (2004), Casassa et al. (2006), Pellicciotti et al. (2008), Vuille et al. (2008), Buttstadt et al. (2009), Rabatel et al., 2011, Rabatel et al., 2013, and MacDonell et al. (2013) performed mass-balance analysis for glaciers in Tierra del Fuego and the Andes Cordillera using the glaciological method, firn cores, and simulations. In several of these examples, Ba time series were compared with different large-scale atmospheric and oceanic indices. However, these studies in general only include individual or few glaciers, observed over limited areas and periods. There thus remains a need for a more comprehensive spatiotemporal analysis of mass-balance trends for the regions discussed, along the Andes Cordillera to the sub-Antarctic islands, and their relationship with global climate indices.

In this study, we analyze 25 glacier Ba time series available for South America (1993–2012) (including the northern (Region 1), central (Region 2), and southern Andes regions (Region 3)), and the sub-Antarctic islands located around the northern tip of the Antarctic Peninsula (Region 4) (Fig. 1), in doing so improving the current spatial and temporal understanding of mass balance trends in these regions. We combine this larger than normal set of glacier estimates of Ba with accumulation area ratio (AAR) methods which include specific routines for estimating under-sampling errors and geographical biases (see Section 2.2). With these data and methods, we aimed to quantify: a) the glacier mass-balance conditions; b) the extent to which the glaciers sampled are out of balance with the present-day climate conditions; and c) the potential climate drivers of these patterns.

In addition to the above effort, we focus on a glacier transect from the tropical latitudes in the northern part of South America to the tip of South America, emphasizing longitudinal gradients and differences in glacier conditions along the Andes Cordillera, and to the sub-Antarctic islands, across the Drake Passage (Fig. 1). Our effort here is aimed at testing hypotheses as to whether sets of glaciers can be considered characteristically together or not, potentially challenging the current opinion that the South American glacier sample is composed of three regions; a northern, central, southern region.

One way of understanding glacier mass-balance and imbalance conditions in a changing climate perspective is to compare Ba time series with different large-scale atmospheric and oceanic indices, for example, the Multivariate El Niño Southern Oscillation (ENSO) (Wolter and Timlin, 2011) and the Pacific Decadal Oscillation (PDO) (Zhang et al., 1997). These atmospheric circulation indices are considered good measures of airflow and moisture transport variability (e.g., Carrasco et al., 2005, Garreaud, 2009, McClung, 2013, López-Moreno et al., 2014) and this variation is considered to be important for glacier conditions (e.g., Francou et al., 2003, Francou et al., 2004, Favier et al., 2004, Vuille et al., 2008, Sagredo and Lowell, 2012, Saltzmann et al., 2013, Veettil et al., 2014, Malmros et al., in review).

Glacier mass loss and the dramatic thinning in the northern Andes may be explained by the higher frequency of El Niño events and changes in their spatial and temporal occurrence together with a warming troposphere (Rabatel et al., 2013). Following this idea, in addition to the mass-balance and out of balance analysis, we evaluate the patterns of temporal and spatial glacier Ba variations using an Empirical Orthogonal Function (EOF) analysis. We combine this with cross-correlations analyses relating Ba patterns from South America and the sub-Antarctic islands to: 1) mesoscale atmospheric-ocean conditions, such as annual ENSO Index and PDO values; and 2) to glacier elevations and aspects, in order to analyze statistical relations and better understanding their linkages.

Section snippets

Observed glaciers

We compiled a dataset of observed annual Ba (kg m 2 yr 1) for 25 glaciers spanning the period 1993 to 2012 for the continental mountain range transect along the Andes Cordillera (6.5°N–45.8°S) to the sub-Antarctic islands (62.7–63.8°S) (Fig. 1). Ba and AAR data originate from the World Glacier Monitoring Service (WGMS 2012, 2013, and earlier issues) database, with additional data from principal investigators, Dyurgerov and Meier (2005), Bahr et al. (2009), and Mernild et al. (2013). Both the Ba

South America and sub-Antarctic islands

The decadal mean time series for the observed glacier Ba between 1993 and 2012 are shown in Fig. 4 for South America and the sub-Antarctic islands, and each of the four individual regions: northern Andes, central Andes, southern Andes, and the sub-Antarctic islands. For South America and the sub-Antarctic islands, the Ba time series show greater mean mass loss for the latest decadal period of − 700 ± 210 kg m 2 yr 1 (2003–2012) (here and below, ± equals the error range) than for the first decadal

Summary and perspectives

Our findings show that decadal Ba was negative on average from 1993 to 2012 for glaciers observed in South America, along the Andes Cordillera, and the sub-Antarctic islands, following the global mean decadal glacier Ba trend estimated by Cogley, 2009, Cogley, 2012. Longitudinally, the decadal Ba from the northern Andes to the sub-Antarctic islands increased on average towards less negative values from 1993 to 2012. On a regional scale, however, Ba trends were shown to be similar for the

Acknowledgments

This study was funded by Chilean FONDECYT (Project: #1140172). We extend a very special thank to World Glacier Monitoring Service (WGMS) for providing us with data from the WGMS database. In addition, we thank the principal investigators of the WGMS network for sharing their observations with the community. The authors have no conflict of interest.

References (80)

  • D.B. Bahr et al.

    The physical basis of glacier volume–area scaling

    J. Geophys. Res.

    (1997)
  • D.B. Bahr et al.

    Sea-level rise from glaciers and ice caps: a lower bound

    Geophys. Res. Lett.

    (2009)
  • R. Barry et al.

    Atmosphere, Weather and Climate

    (2009)
  • J.-M. Beckers et al.

    EOF calculations and data filling from incomplete oceanographic datasets

    J. Atmos. Ocean. Technol.

    (2003)
  • A. Bliss et al.

    A new inventory of mountain glaciers and ice caps for the Antarctic periphery

    Ann. Glaciol.

    (2013)
  • B.B. Booth et al.

    Aerosols implicated as a prime driver of twentieth-century North Atlantic climate variability

    Nature

    (2012)
  • M. Buttstadt et al.

    Mass balance evolution of martial Este Glacier, Tierra del Fuego (Argentina) for the period 1960–2099

    Adv. Geosci.

    (2009)
  • J.F. Carrasco et al.

    Changes of the 0 °C isotherm and the equilibrium line altitude in the central Chile during the last quarter of the 20th century

    Hydrol. Sci. — J. Sci. Hydrol.

    (2005)
  • G. Casassa et al.

    Glacier mass balance data for southern South America (30°S–56°S)

  • J.G. Cogley

    Geodetic and direct mass-balance measurements: comparison and joint analysis

    Ann. Glaciol.

    (2009)
  • J.G. Cogley

    The future of the world's glaciers

  • B.J. Davies et al.

    Accelerating shrinkage of Patagonian glaciers from the Little Ice Age (~ 1870) to 2011

    J. Glaciol.

    (2012)
  • M.B. Dyurgerov

    Data of Glaciological Studies—Reanalysis of Glacier Changes: From the IGY to the IPY, 1960–2008. Publication No. 108

    (2010)
  • M.B. Dyurgerov et al.

    Glaciers and the Changing Earth System: A 2004 Snapshot, Occas

    (2005)
  • M. Dyurgerov et al.

    A new index of glacier area change; a tool for glacier monitoring

    J. Glaciol.

    (2009)
  • F. Escobar et al.

    Variaciones de un glaciar de Montaña en los Andes de Chile Central en las últimas dos décadas

    Bull. Inst. Fr. Etudes Andines (Lima)

    (1995)
  • V. Favier et al.

    Glacier of the outer and inner tropics: a difference behavior but a common response to climatic forcing

    Geophys. Res. Lett.

    (2004)
  • R.L. Fogt et al.

    Understanding the SAM influence on the Southern Pacific ENSO teleconnection

    Clim. Dyn.

    (2011)
  • B. Francou et al.

    Tropical climate change recorded by a glacier in the central Andes during the last decade of the twentieth century: Chacaltaya, Bolivia, 16° S

    J. Geophys. Res.

    (2003)
  • B. Francou et al.

    New evidence for an ENSO impact on low-latitude glaciers: Antizana 15, Andes of Ecuador, 0′28°S

    J. Geophys. Res.

    (2004)
  • A.S. Gardner et al.

    Reconciled estimate of glacier contributions to sea level rise: 2003 to 2009

    Science

    (2013)
  • R.D. Garreaud

    The Andes climate and weather

    Adv. Geosci.

    (2009)
  • C. Georges

    20th-century glacier fluctuations in the tropical Cordillera Blanca, Perú

    Arct. Antarct. Alp. Res.

    (2004)
  • R. Hock et al.

    Mountain glaciers and ice caps around Antarctica make a large sea-level rise contribution

    Geophys. Res. Lett.

    (2009)
  • IPCC

    Observations: cryosphere. Chapter 4

  • IPCC

    Summary of Policy Makers

    (2013)
  • T. Jóhannesson et al.

    Time-scale for adjustment of glaciers to changes in mass balance

    J. Glaciol.

    (1989)
  • A. Kaplan et al.

    Analyses of global sea surface temperatures 1856–1991

    J. Geophys. Res.

    (1998)
  • D.J. Karoly

    Southern hemisphere circulation features associated with El Nino-Southern Oscillation events

    J. Clim.

    (1989)
  • G. Kaser et al.

    Mass balance of glaciers and ice caps: consensus estimates for 1961–2004

    Geophys. Res. Lett.

    (2006)
  • Cited by (34)

    • Snow cover and snow albedo changes in the central Andes of Chile and Argentina from daily MODIS observations (2000–2016)

      2018, Remote Sensing of Environment
      Citation Excerpt :

      During El Niño events, precipitation increases predominantly during the austral winter (Masiokas et al., 2006; McClung, 2013). While El Niño events do influence precipitation amounts, these events shows little or no significant signal in annual mass balance measurements of glaciers located in the central Andes but has been linked to the Pacific Decadal Oscillation (PDO) rather than the ENSO (Mernild et al., 2015a). Along the central Andes, annual accumulation of snow is highest at 4000–5000 m a.s.l., where glacier accumulation zones are also present (Cornwell et al., 2016; Mernild et al., 2016b; Mernild et al., 2016c).

    • Glacial lakes of the Central and Patagonian Andes

      2018, Global and Planetary Change
    • Recent regional climate cooling on the Antarctic Peninsula and associated impacts on the cryosphere

      2017, Science of the Total Environment
      Citation Excerpt :

      The resulting SMB is slightly positive for the period that they analyse (2007–2011), indicating that frontal ablation (combination of iceberg calving and submarine melting at the glacier fronts) is currently the dominant process of mass loss in the SSI (Osmanoğlu et al., 2013, 2014). The fact that the SMB of the SSI, Vega Island and James Ross Island glaciers are currently slightly positive is in agreement with recent findings by Mernild et al. (2015) indicating that the glaciers on the sub-Antarctic islands are only slightly out of balance (4%). This study has an added interest, as it analyses the teleconnections of the observed SMBs.

    View all citing articles on Scopus
    View full text