scholarly journals Improving surface melt estimation over the Antarctic Ice Sheet using deep learning: a proof of concept over the Larsen Ice Shelf

2021 ◽  
Vol 15 (12) ◽  
pp. 5639-5658
Author(s):  
Zhongyang Hu ◽  
Peter Kuipers Munneke ◽  
Stef Lhermitte ◽  
Maaike Izeboud ◽  
Michiel van den Broeke
Keyword(s):  
Deep Learning ◽  
Climate Model ◽  
Ice Sheet ◽  
Katabatic Wind ◽  
Ice Shelf ◽  
Antarctic Ice Sheet ◽  
Surface Melt ◽  
Mlp Model ◽  
The Antarctic ◽  
Larsen Ice Shelf

Abstract. Accurately estimating the surface melt volume of the Antarctic Ice Sheet is challenging and has hitherto relied on climate modeling or observations from satellite remote sensing. Each of these methods has its limitations, especially in regions with high surface melt. This study aims to demonstrate the potential of improving surface melt simulations with a regional climate model by deploying a deep learning model. A deep-learning-based framework has been developed to correct surface melt from the regional atmospheric climate model version 2.3p2 (RACMO2), using meteorological observations from automatic weather stations (AWSs) and surface albedo from satellite imagery. The framework includes three steps: (1) training a deep multilayer perceptron (MLP) model using AWS observations, (2) correcting Moderate Resolution Imaging Spectroradiometer (MODIS) albedo observations, and (3) using these two to correct the RACMO2 surface melt simulations. Using observations from three AWSs at the Larsen B and C ice shelves, Antarctica, cross-validation shows a high accuracy (root-mean-square error of 0.95 mm w.e. d−1, mean absolute error of 0.42 mm w.e. d−1, and a coefficient of determination of 0.95). Moreover, the deep MLP model outperforms conventional machine learning models and a shallow MLP model. When applying the trained deep MLP model over the entire Larsen Ice Shelf, the resulting corrected RACMO2 surface melt shows a better correlation with the AWS observations for two out of three AWSs. However, for one location (AWS 18), the deep MLP model does not show improved agreement with AWS observations; this is likely because surface melt is largely driven by factors (e.g., air temperature, topography, katabatic wind) other than albedo within the corresponding coarse-resolution model pixels. Our study demonstrates the opportunity to improve surface melt simulations using deep learning combined with satellite albedo observations. However, more work is required to refine the method, especially for complicated and heterogeneous terrains.

scholarly journals Improving Surface Melt Estimation over Antarctica Using Deep Learning: A Proof-of-Concept over the Larsen Ice Shelf

2021 ◽  
Author(s):  
Zhongyang Hu ◽  
Peter Kuipers Munneke ◽  
Stef Lhermitte ◽  
Maaike Izeboud ◽  
Michiel van den Broeke
Keyword(s):  
Deep Learning ◽  
Climate Model ◽  
High Surface ◽  
Absolute Error ◽  
Coefficient Of Determination ◽  
Climate Modelling ◽  
Ice Shelf ◽  
Surface Melt ◽  
Mlp Model ◽  
Larsen Ice Shelf

Abstract. Accurately estimating surface melt volume of the Antarctic Ice Sheet is challenging, and has hitherto relied on climate modelling, or on observations from satellite remote sensing. Each of these methods has its limitations, especially in regions with high surface melt. This study aims to demonstrate the potential of improving surface melt simulations by deploying a deep learning model. A deep-learning-based framework has been developed to correct surface melt from the regional atmospheric climate model version 2.3p2 (RACMO2), using meteorological observations from automatic weather stations (AWSs), and surface albedo from satellite imagery. The framework includes three steps: (1) training a deep multilayer perceptron (MLP) model using AWS observations; (2) correcting moderate resolution imaging spectroradiometer (MODIS) albedo observations, and (3) using these two to correct the RACMO2 surface melt simulations. Using observations from three AWSs at the Larsen B and C Ice Shelves, Antarctica, cross-validation shows a high accuracy (root mean square error = 0.95 mm w.e. per day, mean absolute error = 0.42 mm w.e. per day, and coefficient of determination = 0.95). Moreover, the deep MLP model outperforms conventional machine learning models (e.g., random forest regression, XGBoost) and a shallow MLP model. When applying the trained deep MLP model over the entire Larsen Ice Shelf, the resulting, corrected RACMO2 surface melt shows a better correlation with the AWS observations for two out of three AWSs. However, for one location (AWS 18) the deep MLP model does not show improved agreement with AWS observations, likely due to the heterogeneous drivers of melt within the corresponding coarse resolution model pixels. Our study demonstrates the opportunity to improve surface melt simulations using deep learning combined with satellite albedo observations. On the other hand, more work is required to refine the method, especially for complicated and heterogeneous terrains.

scholarly journals Two decades of Antarctic coastal-change revealed by satellite imagery and deep learning

2021 ◽  
Author(s):  
Celia A. Baumhoer ◽  
Andreas Dietz ◽  
Mariel Dirscherl ◽  
Claudia Kuenzer
Keyword(s):  
Deep Learning ◽  
Ice Sheet ◽  
Continuous Mapping ◽  
Coastal Change ◽  
Ice Shelf ◽  
Antarctic Ice Sheet ◽  
Ice Shelves ◽  
Ross Ice Shelf ◽  
Wilkes Land ◽  
The Antarctic

<p>Antarctica’s coastline is constantly changing by moving glacier and ice shelf fronts. The extent of glaciers and ice shelves influences the ice discharge and sea level contribution of the Antarctic Ice Sheet. Therefore, it is crucial to assess where ice shelf areas with strong buttressing forces are lost. So far, those changes have not been assessed for entire Antarctica within comparable time frames.</p><p>We present a framework for circum-Antarctic coastline extraction based on a U-Net architecture. Antarctic coastal-change is calculated by using a deep learning derived coastline for the year 2018 in combination with earlier manual derived coastlines of 1997 and 2009. For the first time, this allows to compare circum-Antarctic changes in glacier and ice shelf front position for the last two decades. We found that the Antarctic Ice Sheet area decreased by -29,618±1,193 km<sup>2</sup> in extent between 1997-2008 and gained an area of 7,108±1,029km<sup>2</sup> between 2009 and 2018. Retreat dominated for the Antarctic Peninsula and West Antarctica and advance for the East Antarctic Ice Sheet over the entire investigation period. The only exception in East Antarctica was Wilkes Land experiencing simultaneous calving front retreat of several glaciers between 2009-2018. Biggest tabular iceberg calving events occurred at Ronne and Ross Ice Shelf within their natural calving cycle between 1997-2008. Future work includes the continuous mapping of Antarctica’s coastal-change on a more frequent temporal scale.  </p>

scholarly journals Automated mapping of the seasonal evolution of surface meltwater and its links to climate on the Amery Ice Shelf, Antarctica

2021 ◽  
Author(s):  
Peter A. Tuckett ◽  
Jeremy C. Ely ◽  
Andrew J. Sole ◽  
James M. Lea ◽  
Stephen J. Livingstone ◽  
...  
Keyword(s):  
Time Series ◽  
Spatial Distribution ◽  
Climate Model ◽  
Google Earth ◽  
Katabatic Wind ◽  
Lake Area ◽  
Ice Shelf ◽  
Amery Ice Shelf ◽  
Surface Melt ◽  
The Antarctic

Abstract. Surface meltwater is widespread around the margin of the Antarctic Ice Sheet and has the potential to influence ice-shelf stability, ice-dynamic processes and ice-albedo feedbacks. Whilst the general spatial distribution of surface meltwater across the Antarctic continent is now relatively well known, our understanding of the seasonal and multi-year evolution of surface meltwater is limited. Attempts to generate robust time series of melt cover have largely been constrained by computational expense or limited ice surface visibility associated with mapping from optical satellite imagery. Here, we implement an existing meltwater detection method alongside a novel method for calculating visibility metrics within Google Earth Engine. This enables us to quantify uncertainty induced by cloud cover and variable image data coverage, allowing us to automatically generate time series of surface melt area over large spatial and temporal scales. We demonstrate our method on the Amery Ice Shelf region of East Antarctica, analysing 4,164 Landsat 7 and 8 optical images between 2005 and 2020. Results show high interannual variability in surface meltwater cover, with mapped cumulative lake area totals ranging from 384 km2 to 3898 km2 per melt season. However, by incorporating image visibility assessments into our results, we estimate that cumulative total lake areas are on average 42 % higher than minimum mapped values, highlighting the importance of accounting for variations in image visibility when mapping lake areas. In a typical melt season, total lake area remains low throughout November and early December, before increasing, on average, by an order of magnitude during the second half of December. Peak lake area most commonly occurs during January, before decreasing during February as lakes freeze over. We show that modelled melt predictions from a regional climate model provides a good indication of lake cover in the Amery region, and that annual lake coverage is strongly associated with phases of the Southern Annular Mode (SAM); surface melt is typically highest in years with a negative austral summer SAM index. Furthermore, we suggest that melt-albedo feedbacks modulate the spatial distribution of meltwater in the region, with the exposure of blue ice from persistent katabatic wind scouring influencing the susceptibility of melt ponding. Results demonstrate how our method could be scaled up to generate a multi-year time series record of surface water extent from optical imagery at a continent-wide scale.

scholarly journals ISMIP6 Antarctica: a multi-model ensemble of the Antarctic ice sheet evolution over the 21<sup>st</sup> century

2020 ◽  
Author(s):  
Helene Seroussi ◽  
Sophie Nowicki ◽  
Antony J. Payne ◽  
Heiko Goelzer ◽  
William H. Lipscomb ◽  
...  
Keyword(s):  
Mass Loss ◽  
Mass Balance ◽  
Sea Level ◽  
Climate Model ◽  
Ice Sheet ◽  
East Antarctica ◽  
Ice Shelf ◽  
Antarctic Ice Sheet ◽  
Surface Mass ◽  
The Antarctic

Abstract. Ice flow models of the Antarctic ice sheet are commonly used to simulate its future evolution in response to different climate scenarios and inform on the mass loss that would contribute to future sea level rise. However, there is currently no consensus on estimated the future mass balance of the ice sheet, primarily because of differences in the representation of physical processes and the forcings employed. This study presents results from 18 simulations from 15 international groups focusing on the evolution of the Antarctic ice sheet during the period 2015–2100, forced with different scenarios from the Coupled Model Intercomparison Project Phase 5 (CMIP5) representative of the spread in climate model results. The contribution of the Antarctic ice sheet in response to increased warming during this period varies between −7.8 and 30.0 cm of Sea Level Equivalent (SLE). The evolution of the West Antarctic Ice Sheet varies widely among models, with an overall mass loss up to 21.0 cm SLE in response to changes in oceanic conditions. East Antarctica mass change varies between −6.5 and 16.5 cm SLE, with a significant increase in surface mass balance outweighing the increased ice discharge under most RCP 8.5 scenario forcings. The inclusion of ice shelf collapse, here assumed to be caused by large amounts of liquid water ponding at the surface of ice shelves, yields an additional mass loss of 8 mm compared to simulations without ice shelf collapse. The largest sources of uncertainty come from the ocean-induced melt rates, the calibration of these melt rates based on oceanic conditions taken outside of ice shelf cavities and the ice sheet dynamic response to these oceanic changes. Results under RCP 2.6 scenario based on two CMIP5 AOGCMs show an overall mass loss of 10 mm SLE compared to simulations done under present-day conditions, with limited mass gain in East Antarctica.

An assessment of basal melting parameterisations for Antarctic ice shelves

2021 ◽  
Author(s):  
Clara Burgard ◽  
Nicolas Jourdain
Keyword(s):  
Climate Model ◽  
Ice Sheet ◽  
Coupled Model ◽  
Ice Shelf ◽  
Antarctic Ice Sheet ◽  
Ice Shelves ◽  
Global Sea Level ◽  
Basal Melting ◽  
The Antarctic

&lt;p&gt;Ocean-induced melting at the base of ice shelves is one of the main drivers of the currently observed mass loss of the Antarctic Ice Sheet. A good understanding of the interaction between ice and ocean at the base of the ice shelves is therefore crucial to understand and project the Antarctic contribution to global sea-level rise.&amp;#160;&lt;/p&gt;&lt;p&gt;Due to the high difficulty to monitor these regions, our understanding of the processes at work beneath ice shelves is limited. Still, several parameterisations of varying complexity have been developed in past decades to describe the ocean-induced sub-shelf melting. These parameterisations can be implemented into standalone ice-sheet models, for example when conducting long-term projections forced with climate model output.&lt;/p&gt;&lt;p&gt;An assessment of the performance of these parameterisations was conducted in an idealised setup (Favier et al, 2019). However, the application of the better-performing parameterisations in a more realistic setup (e.g. Jourdain et al., 2020) has shown that individual adjustments and corrections are needed for each ice shelf.&lt;/p&gt;&lt;p&gt;In this study, we revisit the assessment of the parameterisations, this time in a more realistic setup than previous studies. To do so, we apply the different parameterisations on several ice shelves around Antarctica and compare the resulting melt rates to satellite and oceanographic estimates. Based on this comparison, we will refine the parameters and propose an approach to reduce uncertainties in long-term sub-shelf melting projections.&lt;/p&gt;&lt;p&gt;&lt;em&gt;References&lt;/em&gt;&lt;br&gt;&lt;em&gt;- Favier, L., Jourdain, N. C., Jenkins, A., Merino, N., Durand, G., Gagliardini, O., Gillet-Chaulet, F., and Mathiot, P.: Assessment of sub-shelf melting parameterisations using the ocean&amp;#8211;ice-sheet coupled model NEMO(v3.6)&amp;#8211;Elmer/Ice(v8.3) , Geosci. Model Dev., 12, 2255&amp;#8211;2283, https://doi.org/10.5194/gmd-12-2255-2019, 2019.&amp;#160;&lt;/em&gt;&lt;br&gt;&lt;em&gt;- Jourdain, N. C., Asay-Davis, X., Hattermann, T., Straneo, F., Seroussi, H., Little, C. M., and Nowicki, S.: A protocol for calculating basal melt rates in the ISMIP6 Antarctic ice sheet projections, The Cryosphere, 14, 3111&amp;#8211;3134, https://doi.org/10.5194/tc-14-3111-2020, 2020.&amp;#160;&lt;/em&gt;&lt;/p&gt;

scholarly journals ISMIP6 Antarctica: a multi-model ensemble of the Antarctic ice sheet evolution over the 21st century

2020 ◽  
Author(s):  
Helene Seroussi ◽  
Heiko Goelzer ◽  
Mathieu Morlighem ◽  
Keyword(s):  
Mass Loss ◽  
Mass Balance ◽  
Sea Level ◽  
Climate Model ◽  
Ice Sheet ◽  
East Antarctica ◽  
Ice Shelf ◽  
Antarctic Ice Sheet ◽  
Surface Mass ◽  
The Antarctic

&lt;div&gt; &lt;div&gt; &lt;div&gt; &lt;p&gt;Ice flow models of the Antarctic ice sheet are commonly used to simulate its future evolution in response to differ- ent climate scenarios and inform on the mass loss that would contribute to future sea level rise. However, there is currently no consensus on estimated the future mass balance of the ice sheet, primarily because of differences in the representation of physical processes and the forcings employed. This study presents results from 18 simulations from 15 international groups focusing on the evolution of the Antarctic ice sheet during the period 2015-2100, forced with different scenarios from the Coupled Model Intercomparison Project Phase 5 (CMIP5) representative of the spread in climate model results. The contribution of the Antarctic ice sheet in response to increased warming during this period varies between -7.8 and 30.0 cm of Sea Level Equivalent (SLE). The evolution of the West Antarctic Ice Sheet varies widely among models, with an overall mass loss up to 21.0 cm SLE in response to changes in oceanic conditions. East Antarctica mass change varies between -6.5 and 16.5 cm SLE, with a significant increase in surface mass balance outweighing the increased ice discharge under most RCP 8.5 scenario forcings. The inclusion of ice shelf collapse, here assumed to be caused by large amounts of liquid water ponding at the surface of ice shelves, yields an additional mass loss of 8 mm compared to simulations without ice shelf collapse. The largest sources of uncertainty come from the ocean-induced melt rates, the calibration of these melt rates based on oceanic conditions taken outside of ice shelf cavities and the ice sheet dynamic response to these oceanic changes. Results under RCP 2.6 scenario based on two CMIP5 AOGCMs show an overall mass loss of 10 mm SLE compared to simulations done under present-day conditions, with limited mass gain in East Antarctica.&lt;/p&gt; &lt;/div&gt; &lt;/div&gt; &lt;/div&gt;

scholarly journals Spatial distribution of net surface accumulation on the Antarctic ice sheet

2000 ◽  
Vol 31 ◽  
pp. 171-178 ◽  
Author(s):  
M. B. Glovinetto ◽  
H. J. Zwally
Keyword(s):  
Spatial Distribution ◽  
Field Data ◽  
Ice Sheet ◽  
Regional Assessment ◽  
Ice Shelf ◽  
Antarctic Ice Sheet ◽  
Drainage Systems ◽  
Continental Scale ◽  
The Antarctic ◽  
Larsen Ice Shelf

AbstractAn isopleth map showing the spatial distribution of net mass accumulation at the surface on the Antarctic ice sheet, excluding Graham Land, the Larsen Ice Shelf and eastern Palmer Land, is produced based on field data from approximately 2000 sites. A database of accumulation values for 5365 gridpoint locations with 50 km spacing is interpolated from the isopleth map, giving a bulk accumulation of 2151 Gt a–1 and a mean of 159 kg m–2 a–1 for an area of 13.53 × 106 km2. Following the implementation of deflation and ablation adjustments applicable to sectors of the coastal zone, the accumulation values are reduced to 2020 Gt a–1 and 149 kg m–2 a–1. The new accumulation distribution is compared with another recent distribution, which was based on essentially the same field data using different analysis and interpolation criteria. Differences between the distributions are assessed using residuals for the 50 km gridpoint locations and by comparing average accumulation values for 24 drainage systems. The assessment based on residuals indicates that the two distributions show patterns of accumulation that are coherent at the continental scale, a shared attribute underscored by a small mean residual value of 6 kg m–2 a–1 (a difference of <4%). However, the regional assessment based on average accumulation values for the drainage systems shows differences that are larger than the assessment error (>22%) for six systems that collectively comprise approximately 4/10 of the ice-sheet area and 3/10 of the accumulation.

Antarctic ice sheet response to upper-bound scenarios

2021 ◽  
Author(s):  
Sainan Sun ◽  
Frank Pattyn
Keyword(s):  
Sea Level Rise ◽  
Sea Level ◽  
Upper Bound ◽  
Ice Sheet ◽  
Climate Forcing ◽  
Ice Shelf ◽  
Antarctic Ice Sheet ◽  
Ice Shelves ◽  
Basal Sliding ◽  
The Antarctic

&lt;p&gt;Mass loss of the Antarctic ice sheet contributes the largest uncertainty of future sea-level rise projections. Ice-sheet model predictions are limited by uncertainties in climate forcing and poor understanding of processes such as ice viscosity. The Antarctic BUttressing Model Intercomparison Project (ABUMIP) has investigated the 'end-member' scenario, i.e., a total and sustained removal of buttressing from all Antarctic ice shelves, which can be regarded as the upper-bound physical possible, but implausible contribution of sea-level rise due to ice-shelf loss. In this study, we add successive layers of &amp;#8216;realism&amp;#8217; to the ABUMIP scenario by considering sustained regional ice-shelf collapse and by introducing ice-shelf regrowth after collapse with the inclusion of ice-sheet and ice-shelf damage (Sun et al., 2017). Ice shelf regrowth has the ability to stabilize grounding lines, while ice shelf damage may reinforce ice loss. In combination with uncertainties from basal sliding and ice rheology, a more realistic physical upperbound to ice loss is sought. Results are compared in the light of other proposed mechanisms, such as MICI due to ice cliff collapse.&lt;/p&gt;

scholarly journals Antarctic Ice-Sheet Volume at 18000 Years B.P. and Holocene Sea-Level Changes at the West Antarctic Margin

1979 ◽  
Vol 24 (90) ◽  
pp. 213-230 ◽  
Author(s):  
Craig S. Lingle ◽  
James A. Clark
Keyword(s):  
Sea Level Rise ◽  
Sea Level ◽  
Ice Sheet ◽  
Sea Level Changes ◽  
Ice Shelf ◽  
Relative Sea Level ◽  
Antarctic Ice Sheet ◽  
West Antarctic Ice Sheet ◽  
West Antarctic ◽  
The Antarctic

AbstractThe Antarctic ice sheet has been reconstructed at 18000 years b.p. by Hughes and others (in press) using an ice-flow model. The volume of the portion of this reconstruction which contributed to a rise of post-glacial eustatic sea-level has been calculated and found to be (9.8±1.5) × 106 km3. This volume is equivalent to 25±4 m of eustatic sea-level rise, defined as the volume of water added to the ocean divided by ocean area. The total volume of the reconstructed Antarctic ice sheet was found to be (37±6) × 106 km3. If the results of Hughes and others are correct, Antarctica was the second largest contributor to post-glacial eustatic sea-level rise after the Laurentide ice sheet. The Farrell and Clark (1976) model for computation of the relative sea-level changes caused by changes in ice and water loading on a visco-elastic Earth has been applied to the ice-sheet reconstruction, and the results have been combined with the changes in relative sea-level caused by Northern Hemisphere deglaciation as previously calculated by Clark and others (1978). Three families of curves have been compiled, showing calculated relative sea-level change at different times near the margin of the possibly unstable West Antarctic ice sheet in the Ross Sea, Pine Island Bay, and the Weddell Sea. The curves suggest that the West Antarctic ice sheet remained grounded to the edge of the continental shelf until c. 13000 years b.p., when the rate of sea-level rise due to northern ice disintegration became sufficient to dominate emergence near the margin predicted otherwise to have been caused by shrinkage of the Antarctic ice mass. In addition, the curves suggest that falling relative sea-levels played a significant role in slowing and, perhaps, reversing retreat when grounding lines approached their present positions in the Ross and Weddell Seas. A predicted fall of relative sea-level beneath the central Ross Ice Shelf of as much as 23 m during the past 2000 years is found to be compatible with recent field evidence that the ice shelf is thickening in the south-east quadrant.

scholarly journals Surface melt magnitude retrieval over Ross Ice Shelf, Antarctica using coupled MODIS near-IR and thermal satellite measurements

2009 ◽  
Vol 3 (3) ◽  
pp. 1069-1107 ◽  
Author(s):  
D. J. Lampkin ◽  
C. C. Karmosky
Keyword(s):  
Meteorological Data ◽  
Temporal Variations ◽  
Katabatic Wind ◽  
Ice Shelf ◽  
Water Fraction ◽  
Ross Ice Shelf ◽  
Near Ir ◽  
Coarse Spatial Resolution ◽  
Surface Melt ◽  
The Antarctic

Abstract. Surface melt has been increasing over recent years, especially over the Antarctic Peninsula, contributing to disintegration of shelves such as Larsen. Unfortunately, we are not realistically able to quantify surface snowmelt from ground-based methods because there is sparse coverage of automatic weather stations. Satellite based assessments of melt from passive microwave systems are limited in that they only provide an indication of melt occurrence and have coarse spatial resolution. An algorithm was developed to retrieve surface melt magnitude using coupled near-IR/thermal surface measurements from MODIS were calibrated by estimates of liquid water fraction (LWF) in the upper 1 cm of the firn derived from a one-dimensional physical snowmelt model (SNTHERM89). For the modeling phase of this study, SNTHERM89 was forced by hourly meteorological data from automatic weather station data at reference sites spanning a range of melt conditions across the Ross Ice Shelf during a relatively intense melt season (2002). Effective melt magnitude or LWF<eff> were derived for satellite composite periods covering the Antarctic summer months at a 4 km resolution over the entire Ross Ice Shelf, ranging from 0–0.5% LWF<eff> in early December to areas along the coast with as much as 1% LWF<eff> during the time of peak surface melt. Spatial and temporal variations in the magnitude of surface melt are related to both katabatic wind strength and advection during onshore flow.

Sign in / Sign up

Export Citation Format

Share Document