Search

Sea level contributions from the Greenland Ice Sheet are influenced by the rapid changes in glacial terminus positions. The documentation of these evolving calving front positions, for which satellite imagery forms the basis, is therefore important. However, the manual delineation of these calving fronts is time consuming, which limits the availability of these data across a wide spatial and temporal range. Automated methods face challenges that include the handling of clouds, illumination differences, sea ice mélange, and Landsat 7 scan line corrector errors. To address these needs, we develop the Calving Front Machine (CALFIN), an automated method for extracting calving fronts from satellite images of marine-terminating glaciers, using neural networks. The results are often indistinguishable from manually curated fronts, deviating by on average 86.76 ± 1.43 m from the measured front. Landsat imagery from 1972 to 2019 is used to generate 22 678 calving front lines across 66 Greenlandic glaciers. This improves on the state of the art in terms of the spatiotemporal coverage and accuracy of its outputs and is validated through a comprehensive intercomparison with existing studies. The current implementation offers a new opportunity to explore subseasonal and regional trends on the extent of Greenland's margins and supplies new constraints for simulations of the evolution of the mass balance of the Greenland Ice Sheet and its contributions to future sea level rise.

Author(s): Mohajerani, Y; Velicogna, I; Rignot, E | Abstract: We develop regionally optimized Gravity Recovery and Climate Experiment (GRACE) solutions to evaluate the mass balance of the drainage basins of Amery Ice Shelf, East Antarctica, and Getz Ice Shelf, West Antarctica. We find that the Amery region is near balance, while the Getz region is rapidly losing mass. We compare the results with the mass budget method (MBM) combining ice discharge along the periphery with surface mass balance derived from three regional climate models: (1) Regional Atmospheric Climate Model (RACMO) 2.3p1 and (2) 2.3p2, and (3) Modele Atmospherique Regional 3.6.41. For Amery, MBM/RACMO2.3p1 agrees with GRACE, while MBM/RACMO2.3p2 and MBM/MAR3.6.41 suggest a positive mass balance. For Getz, all estimates agree with a mass loss and the GRACE results are robust to uncertainties in glacial isostatic adjustment derived from an ensemble 128,000 forward models. Over the period April 2002 to November 2015, the mass loss of the Getz drainage basin is 22.9n±n10.9nGt/yr with an acceleration of 1.6n±n0.9nGt/yr2.

Delineating the grounding line of marine-terminating glaciers—where ice starts to become afloat in ocean waters—is crucial for measuring and understanding ice sheet mass balance, glacier dynamics, and their contributions to sea level rise. This task has been previously done using time-consuming, mostly-manual digitizations of differential interferometric synthetic-aperture radar interferograms by human experts. This approach is no longer viable with a fast-growing set of satellite observations and the need to establish time series over entire continents with quantified uncertainties. We present a fully-convolutional neural network with parallel atrous convolutional layers and asymmetric encoder/decoder components that automatically delineates grounding lines at a large scale, efficiently, and accompanied by uncertainty estimates. Our procedure detects grounding lines within 232 m in 100-m posting interferograms, which is comparable to the performance achieved by human experts. We also find value in the machine learning approach in situations that even challenge human experts. We use this approach to map the tidal-induced variability in grounding line position around Antarctica in 22,935 interferograms from year 2018. Along the Getz Ice Shelf, in West Antarctica, we demonstrate that grounding zones are one order magnitude (13.3 ± 3.9) wider than expected from hydrostatic equilibrium, which justifies the need to map grounding lines repeatedly and comprehensively to inform numerical models.

We present Calving Front Machine (CALFIN), an automated method for extracting calving fronts from satellite images of marine-terminating glaciers. The results use Landsat imagery from 1972 to 2019 to generate 22 678 calving front lines across 66 Greenlandic glaciers. The method uses deep learning, and builds on existing work by Mohajerani et al., Zhang et al., and Baumhoer et al. Additional post-processing techniques allow for accurate segmentation of imagery into Shapefile outputs. This method is uniquely robust to the impact of clouds, illumination differences, ice mélange, and Landsat-7 Scan Line Corrector errors. CALFIN provides improvements on the current state of the art. A model inter-comparison is performed to evaluate performance against existing methodologies. CALFIN's ability to generalize to SAR imagery is also evaluated. CALFIN's fronts are often indistinguishable from manually-curated fronts, deviating by 2.25 pixels (86.76 meters) from the true front on a diverse set of 162 testing images. The current implementation offers a new opportunity to explore sub-seasonal trends on the extent of Greenland's margins, and supplies new constraints for simulations of the evolution of the mass balance of the Greenland Ice Sheet and its contributions to future sea level rise.

International audience; We examine data continuity between the Gravity Recovery and Climate Experiment (GRACE) and GRACE Follow‐On (FO) missions over Greenland and Antarctica using independent data from the mass budget method, which calculates the difference between ice sheet surface mass balance and ice discharge at the periphery. For both ice sheets, we find consistent GRACE/GRACE‐FO time series across the data gap, at the continental and regional scales, and the data gap is confidently filled with mass budget method data. In Greenland, the GRACE‐FO data reveal an exceptional summer loss of 600 Gt in 2019 following two cold summers. In Antarctica, ongoing high mass losses in the Amundsen Sea Embayment of West Antarctica, the Antarctic Peninsula, and Wilkes Land in East Antarctica cumulate to 2130, 560, and 370 Gt, respectively, since 2002. A cumulative mass gain of 980 Gt in Queen Maud Land since 2009, however, led to a pause in the acceleration in mass loss from Antarctica after 2016.

We present Calving Front Machine (CALFIN), an automated method for extracting calving fronts from satellite imagery. We generate results for 66 glaciers along East/West Greenland from 1972 to 2019. We output these results as a dataset, and provide new constraints on glacial evolution over the time period. This method is uniquely robust to clouds, illumination differences, ice mélange, and Landsat-7 Scan Line Corrector errors. The current implementation offers a new opportunity to explore previous trends, and validate existing models moving forward.This method utilizes deep learning, in the form of the Google DeeplabV3+ Xception derived CALFIN Neural Network. This approach builds on existing work by Mohajerani et al., Zhang et al., and Baumhoer et al. Additional post-processing techniques allow our method to achieve accurate and useful segmentation of raw images into Shapefile outputs. We achieve are often indistinguishable from the manually-curated fronts, deviating from such test data by 1 pixel (about 80 meters) or less XXX% of the time across 162 test images.CALFIN excels among the current state of the art. We show this by performing a model inter-comparison to evaluate CALFIN's performance against existing methodologies. We also showcase CALFIN's ability to generalize to SAR and MODIS imagery. We achieve a mean error of 2.25 pixels (86.76 meters) from the true front on a diverse set of 162 testing images.

The Antarctic Ice Sheet is an important indicator of climate change and driver of sea-level rise. Here we combine satellite observations of its changing volume, flow and gravitational attraction with modelling of its surface mass balance to show that it lost 2,720 ± 1,390 billion tonnes of ice between 1992 and 2017, which corresponds to an increase in mean sea level of 7.6 ± 3.9 millimetres (errors are one standard deviation). Over this period, ocean-driven melting has caused rates of ice loss from West Antarctica to increase from 53 ± 29 billion to 159 ± 26 billion tonnes per year; ice-shelf collapse has increased the rate of ice loss from the Antarctic Peninsula from 7 ± 13 billion to 33 ± 16 billion tonnes per year. We find large variations in and among model estimates of surface mass balance and glacial isostatic adjustment for East Antarctica, with its average rate of mass gain over the period 1992–2017 (5 ± 46 billion tonnes per year) being the least certain.

Gravity-based estimates of mass change have been extended by the recently launched GRACE Follow-On Satellites. The satellite record, combined with regional climate models, reveals that the Greenland Ice Sheet had lower mass loss in 2017–2018, only to return to a record-breaking mass loss in the summer of 2019.

The continuous and precise mapping of glacier calving fronts is essential for monitoring and understanding rapid glacier changes in Antarctica and Greenland, which have the potential for significant sea level rise within the current century. This effort has been mostly restricted to the slow and painstaking manual digitalization of the calving front positions in thousands of satellite imagery products. Here, we have developed a machine learning toolkit to robustly and automatically detect glacier calving front margins in satellite imagery. The toolkit is based on semantic image segmentation using Convolutional Neural Networks (CNN) with a modified U-Net architecture to isolate the calving fronts from satellite images after having been trained with a dataset of images and their corresponding manually-determined calving fronts. As a case study we train our neural network on a varied set Landsat images with lowered resolutions from Jakobshavn, Sverdrup, and Kangerlussuaq glaciers, Greenland and test the results on novel images from Helheim glacier, Greenland to evaluate the performance of the approach. The neural network is able to identify the calving front in new images with a mean deviation of 96.3 m from the true fronts, equivalent to 1.97 pixels on average, while the corresponding error for manually-determined fronts on the same resolution images is 92.5 m. We find that the trained neural network significantly outperforms common edge detection techniques, and can be used to continuously map out calving-ice fronts with a variety of data products.

Changes in extreme temperature and precipitation may give some of the largest significant societal and ecological impacts. For changes in the magnitude of extreme temperature and precipitation over India, we used a statistical model of generalized extreme value (GEV) distribution. The GEV statistical distribution is a time-dependent distribution with different time scales of variability bounded by a precipitation, maximum (Tmax), and minimum (Tmin) temperature extremes and also assessed their possibility changes are evaluated and quantified over India is presented. The GEV-based method is applied on both precipitation and temperature extremes over India during the 20th and 21st centuries using multiple coupled climate models taking an interest in the Coupled Model Intercomparison Project Phase 5 (CMIP5) and observational datasets. The regional means of historical warm extreme temperatures are 34.89, 36.42, and 38.14 °C for three different (10, 20, and 50-year) periods, respectively; whereas the cold extreme mean temperatures are 7.75, 4.19, and −1.57 °C. It indicates that 20th century cold extreme temperatures have relatively larger variations than the warm extremes. As for the future, the CMIP5 models of warm extreme regional mean values increase from 0.33 to 0.75 °C in all return periods (10-, 20-, and 50-year periods), while in the case of cold extreme means values vary between 0.58 and 2.29 °C. In the future, cold extreme values have a larger increasing rate over the northwest, northeast, some parts of north-central, and Inter Peninsula regions. The CRU precipitation extremes are larger than the historical extreme precipitation in all three (10, 20, and 50-year) return-periods.

ArticlePublished: 10 December 2019This is an unedited manuscript that has been accepted for publication. Nature Research are providing this early version of the manuscript as a service to our customers. The manuscript will undergo copyediting, typesetting and a proof review before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers apply.Mass balance of the Greenland Ice Sheet from 1992 to 2018The IMBIE TeamNature (2019)Cite this article6914 Accesses1410 AltmetricMetricsdetailsAbstractIn recent decades, the Greenland Ice Sheet has been a major contributor to global sea-level rise1,2, and it is expected to be so in the future3. Although increases in glacier flow4–6 and surface melting7–9 have been driven by oceanic10–12 and atmospheric13,14 warming, the degree and trajectory of today’s imbalance remain uncertain. Here we compare and combine 26 individual satellite measurements of changes in the ice sheet’s volume, flow and gravitational potential to produce a reconciled estimate of its mass balance. Although the ice sheet was close to a state of balance in the 1990s, annual losses have risen since then, peaking at 335 ± 62 billion tonnes per year in 2011. In all, Greenland lost 3,800 ± 339 billion tonnes of ice between 1992 and 2018, causing the mean sea level to rise by 10.6 ± 0.9 millimetres. Using three regional climate models, we show that reduced surface mass balance has driven 1,971 ± 555 billion tonnes (52%) of the ice loss owing to increased meltwater runoff. The remaining 1,827 ± 538 billion tonnes (48%) of ice loss was due to increased glacier discharge, which rose from 41 ± 37 billion tonnes per year in the 1990s to 87 ± 25 billion tonnes per year since then. Between 2013 and 2017, the total rate of ice loss slowed to 217 ± 32 billion tonnes per year, on average, as atmospheric circulation favoured cooler conditions15 and as ocean temperatures fell at the terminus of Jakobshavn Isbræ16. Cumulative ice losses from Greenland as a whole have been close to the IPCC’s predicted rates for their high-end climate warming scenario17, which forecast an additional 50 to 120 millimetres of global sea-level rise by 2100 when compared to their central estimate.