Horst Machguth

Strong spatial variability of snow accumulation observed with helicopter-borne GPR on two adjacent Alpine glaciers

The spatial distribution of snow accumulation plays an essential role for the mass balancedistribution on alpine glaciers. Traditional point measurements (snow pits and-probes) are labour intensive and interpolation in-between the points causes uncertainties.Airborne radar measurements have already been used for snow mapping innon-glacierized terrain, but not on Alpine glaciers. To enhance our understanding ofthe spatial distribution of accumulation and pin down reasons for observed variations,we have conducted high-resolution helicopter-borne radar measurements on the temperatedglacier Findel and neighbouring Adler Glacier in southwestern Switzerland.The radar sensor was mounted underneath a helicopter and operated at a centre frequencyof 500 MHz with a bandwidth of 400 MHz. The results were validated withextensive ground-based profiling of the snow cover. The radar data allows a clearrecognition of the snow cover (6% of the total profile length of 10 km did not allowinterpretation due to missing or disturbed layering) and agreed well with the groundbased measurements (R2 = 0.85). Reduced accumulation has been observed in allcrevassed zones. Statistical analysis of the correlation between observed accumulationand terrain characteristics have been performed in a GIS environment, revealing differingaccumulation patterns: On the lower part of Findel Glacier accumulation showsa clear altitudinal trend, while the upper part is dominated by strongly varying snowdepth without an altitudinal trend. The accumulation characteristic on Adler Glacieris similar to the upper part of Findel Glacier, but despite of their close vicinity, accumulationis reduced by 40% compared to the same elevation on Findel Glacier. Thisstudy reveals a large potential of helicopter-borne snow profiling for measurements ofaccumulation distribution on alpine glaciers.

Distributed glacier mass-balance modelling as an important component of modern multi-level glacier monitoring

Abstract Modern concepts of worldwide glacier monitoring include numerical models for (1) interconnecting the different levels of observations (local mass balance, representative length change, glacier inventories for global coverage) and (2) extrapolations in space (coupling with climate models) and time (backward and forward). In this context, one important new tool is distributed mass-balance modelling in complex mountain topography. This approach builds on simplified energy-balance models and can be applied for investigating the spatio-temporal representativity of the few mass-balance measurements, for estimating balance values at the tongue of unmeasured glaciers in order to derive long-term average balance values from a great number of glaciers with known length change, and for assessing special effects such as the influence of Sahara dust falls on the albedo and mass balance or autocorrelation effects due to surface darkening of glaciers with strongly negative balances. Experience from first model runs in the Swiss Alps and from applications to the extreme conditions in summer 2003 provides evidence about the usefulness of this approach for glacier monitoring and analysis of glacier changes in high-mountain regions. The main difficulties concern the spatial variability of the input parameters (e.g. precipitation, snow cover and surface albedo) and the uncertainties in the parameterizations of the components of the energy balance. Field measurements remain essential to tie the models to real ground conditions.

Calculating distributed glacier mass balance for the Swiss Alps from regional climate model output: A methodical description and interpretation of the results

[1] This study aims at giving a methodical description of the use of gridded output from a regional climate model (RCM) for the calculation of glacier mass balance distribution for the perimeter of the Swiss Alps. The mass balance model runs at daily steps and 100 m spatial resolution, while the regional model (REMO) RCM provides daily grids (� 18 km resolution) of dynamically downscaled reanalysis data. A combination of interpolation techniques and simple subgrid parameterizations is applied to bridge the gap in spatial resolution and to obtain daily input fields of air temperature, global radiation, and precipitation. Interpolation schemes are a key element and thus we test different interpolators. For validation, computed mass balances are compared to stake measurements and time series (1979–2003) of observed mass balance. The meteorological input fields are compared to measurements at weather stations. The applied inverse distance weighting introduces systematic biases due to spatial autocorrelation, whereas thin plate splines preserve the characteristics of the RCM output. While summer melt at point locations on several glaciers is well reproduced by the model, accumulation is mostly underestimated. These systematic shifts are correlated to biases of the meteorological input fields. Time series of mass balance obtained from the model run agree well with observed time series. We conclude that the gap in spatial resolution is not a major drawback, given that interpolators and parameterizations are selected upon detailed considerations. Biases in RCM precipitation are a major source for the observed underestimations in mass balance and have to be corrected prior to operational use of the presented approach.

Fast shrinkage of tropical glaciers in Colombia

Abstract As a consequence of ongoing atmospheric temperature rise, tropical glaciers belong to the unique and threatened ecosystems on Earth, as defined by the Intergovernmental Panel on Climate Change (Houghton and others, 2001). Worldwide glacier monitoring, especially as part of the Global Climate Observing System (GCOS), includes the systematic collection of data on such perennial surface ice masses. Several peaks in the sierras of Colombia have lost their glacier cover during recent decades. Today, high-altitude glaciers still exist in Sierra Nevada de Santa Marta, in Sierra Nevada del Cocuy and on the volcanoes of Nevados del Ruiz, de Santa Isabel, del Tolima and del Huila. Comparison of reconstructions of maximum glacier area extent during the Little Ice Age with more recent information from aerial photographs and satellite images clearly documents a fast-shrinking tendency and potential disappearance of the remaining glaciers within the next few decades. In the past 50 years, Colombian glaciers have lost 5 0% or more of their area. Glacier shrinkage has continued to be strong in the last 15 years, with a loss of 10−50% of the glacier area. The relationship between fast glacier retreat and local, regional and global climate change is now being investigated. Preliminary analyses indicate that the temperature rise of roughly 1°C in the last 30 years recorded at high-altitude meteorological stations exerts a primary control on glacier retreat. The investigations on the Colombian glaciers thus corroborate earlier findings concerning the high sensitivity of glaciers in the wet inner tropics to temperature rise. To improve understanding of fast glacier retreat in Colombia, a modern monitoring network has been established according to the multilevel strategy of the Global Terrestrial Network for Glaciers (GTN-G) within GCOS. The observations are also contributions to continued assessments of hazards from the glacier-covered volcanoes and to integrated global change research in mountain biosphere reserves.

Modelling glacier-bed overdeepenings and possible future lakes for the glaciers in the Himalaya—Karakoram region

Abstract Surface digital elevation models (DEMs) and slope-related estimates of glacier thickness enable modelling of glacier-bed topographies over large ice-covered areas. Due to the erosive power of glaciers, such bed topographies can contain numerous overdeepenings, which when exposed following glacier retreat may fill with water and form new lakes. In this study, the bed overdeepenings for ~28 000 glaciers (40 775 km2) of the Himalaya-Karakoram region are modelled using GlabTop2 (Glacier Bed Topography model version 2), in which ice thickness is inferred from surface slope by parameterizing basal shear stress as a function of elevation range for each glacier. The modelled ice thicknesses are uncertain (±30%), but spatial patterns of ice thickness and bed elevation primarily depend on surface slopes as derived from the DEM and, hence, are more robust. About 16 000 overdeepenings larger than 104m2 were detected in the modelled glacier beds, covering an area of ~2200 km2 and having a volume of ~120km3 (3-4% of present-day glacier volume). About 5000 of these overdeepenings (1800 km2) have a volume larger than 106m3. The results presented here are useful for anticipating landscape evolution and potential future lake formation with associated opportunities (tourism, hydropower) and risks (lake outbursts).

Comparing three different methods to model scenarios of future glacier change in the Swiss Alps

Abstract Ongoing atmospheric warming causes rapid shrinking of glaciers in the European Alps, with a high chance of their near-complete disappearance by the end of the 21st century. Here we present a comparison of three independent approaches to model the possible evolution of the glaciers in the Swiss Alps over the 21st century. The models have different levels of complexity, work at a regional scale and are forced with three scenarios of temperature increase (low, moderate, high). The moderate climate scenario gives an increase in air temperature of ∼2°C and ∼4°C for the two scenario periods 2021-50 and 2070-99, respectively, resulting in an area loss of 60-80% by 2100. In reality, the shrinkage could be even faster, as the observed mean annual thickness loss is already stronger than the modelled one. The three approaches lead to rather similar results with respect to the overall long-term evolution. The choice of climate scenarios produces the largest spread (∼40%) in the final area loss, while the uncertainty in present-day ice-thickness estimation causes about half this spread.

The future sea-level rise contribution of Greenland’s glaciers and ice caps

We calculate the future sea-level rise contribution from the surface mass balance of all of Greenland?s glaciers and ice caps (GICs, ?90?000?km2) using a simplified energy balance model which is driven by three future climate scenarios from the regional climate models HIRHAM5, RACMO2 and MAR. Glacier extent and surface elevation are modified during the mass balance model runs according to a glacier retreat parameterization. Mass balance and glacier surface change are both calculated on a 250?m resolution digital elevation model yielding a high level of detail and ensuring that important feedback mechanisms are considered. The mass loss of all GICs by 2098 is calculated to be 2016???129?Gt (HIRHAM5 forcing), 2584???109?Gt (RACMO2) and 3907???108?Gt (MAR). This corresponds to a total contribution to sea-level rise of 5.8???0.4, 7.4???0.3 and 11.2???0.3?mm, respectively. Sensitivity experiments suggest that mass loss could be higher by 20?30% if a strong lowering of the surface albedo were to take place in the future. It is shown that the sea-level rise contribution from the north-easterly regions of Greenland is reduced by increasing precipitation while mass loss in the southern half of Greenland is dominated by steadily decreasing summer mass balances. In addition we observe glaciers in the north-eastern part of Greenland changing their characteristics towards greater activity and mass turnover.

The Swiss Alpine glaciers' response to the global '2 C air temperature target'

While there is general consensus that observed global mean air temperature has increased during the past few decades and will very likely continue to rise in the coming decades, the assessment of the effective impacts of increased global mean air temperature on a specific regional-scale system remains highly challenging. This study takes up the widely discussed concept of limiting global mean temperature to a certain target value, like the so-called 2 °C target, to assess the related impacts on the Swiss Alpine glaciers. A model setup is introduced that uses and combines homogenized long-term meteorological observations and three ensembles of transient gridded Regional Climate Model simulations to drive a distributed glacier mass balance model under a (regionalized) global 2 °C target scenario. 101 glaciers are analyzed representing about 50% of the glacierized area and 75% of the ice volume in Switzerland. In our study, the warming over Switzerland, which corresponds to the global 2 °C target is met around 2030, 2045 and 2055 (depending on the ensemble) for Switzerland, and all glaciers have fully adjusted to the new climate conditions at around 2150. By this time and relative to the year 2000, the glacierized area and volume are both decreased to about 35% and 20%, respectively, and glacier-based runoff is reduced by about 70%.

A tipping point in refreezing accelerates mass loss of Greenland’s glaciers and ice caps

Melting of the Greenland ice sheet (GrIS) and its peripheral glaciers and ice caps (GICs) contributes about 43% to contemporary sea level rise. While patterns of GrIS mass loss are well studied, the spatial and temporal evolution of GICs mass loss and the acting processes have remained unclear. Here we use a novel, 1 km surface mass balance product, evaluated against in situ and remote sensing data, to identify 1997 (±5 years) as a tipping point for GICs mass balance. That year marks the onset of a rapid deterioration in the capacity of the GICs firn to refreeze meltwater. Consequently, GICs runoff increases 65% faster than meltwater production, tripling the post-1997 mass loss to 36±16 Gt−1, or ∼14% of the Greenland total. In sharp contrast, the extensive inland firn of the GrIS retains most of its refreezing capacity for now, buffering 22% of the increased meltwater production. This underlines the very different response of the GICs and GrIS to atmospheric warming.

Mass Balance Re-analysis of Findelengletscher, Switzerland; Benefits of Extensive Snow Accumulation Measurements

A re-analysis is presented here of a 10-year mass balance series at Findelengletscher, a temperate mountain glacier in Switzerland. Calculating glacier-wide mass balance from the set of glaciological point balance observations using conventional approaches, such as the profile or contour method, resulted in significant deviations from the reference value given by the geodetic mass change over a five-year period. This is attributed to the sparsity of observations at high elevations and to the inability of the evaluation schemes to adequately estimate accumulation in unmeasured areas. However, measurements of winter mass balance were available for large parts of the study period from snow probings and density pits. Complementary surveys by helicopter-borne ground-penetrating radar (GPR) were conducted in three consecutive years. The complete set of seasonal observations was assimilated using a distributed mass balance model. This model-based extrapolation revealed a substantial mass loss at Findelengletscher of -0.43m w.e. a^-1 between 2004 and 2014, while the loss was less pronounced for its former tributary, Adlergletscher (-0.30m w.e. a^-1). For both glaciers, the resulting time series were within the uncertainty bounds of the geodetic mass change. We show that the model benefited strongly from the ability to integrate seasonal observations. If no winter mass balance measurements were available and snow cover was represented by a linear precipitation gradient, the geodetic mass balance was not matched. If winter balance measurements by snow probings and snow density pits were taken into account, the model performance was substantially improved but still showed a significant bias relative to the geodetic mass change. Thus the excellent agreement of the model-based extrapolation with the geodetic mass change was owed to an adequate representation of winter accumulation distribution by means of extensive GPR measurements.