scholarly journals Spaceborne GNSS-R for Sea Ice Classification Using Machine Learning Classifiers

2021 ◽  
Vol 13 (22) ◽  
pp. 4577
Author(s):  
Yongchao Zhu ◽  
Tingye Tao ◽  
Jiangyang Li ◽  
Kegen Yu ◽  
Lei Wang ◽  
...  
Keyword(s):  
Machine Learning ◽  
Sea Ice ◽  
Open Water ◽  
Arctic Sea Ice ◽  
Support Vector ◽  
First Year ◽  
Training Set ◽  
Ice Coverage ◽  
Input Variables ◽  
Microwave Imager

The knowledge of Arctic Sea ice coverage is of particular importance in studies of climate change. This study develops a new sea ice classification approach based on machine learning (ML) classifiers through analyzing spaceborne GNSS-R features derived from the TechDemoSat-1 (TDS-1) data collected over open water (OW), first-year ice (FYI), and multi-year ice (MYI). A total of eight features extracted from GNSS-R observables collected in five months are applied to classify OW, FYI, and MYI using the ML classifiers of random forest (RF) and support vector machine (SVM) in a two-step strategy. Firstly, randomly selected 30% of samples of the whole dataset are used as a training set to build classifiers for discriminating OW from sea ice. The performance is evaluated using the remaining 70% of samples through validating with the sea ice type from the Special Sensor Microwave Imager Sounder (SSMIS) data provided by the Ocean and Sea Ice Satellite Application Facility (OSISAF). The overall accuracy of RF and SVM classifiers are 98.83% and 98.60% respectively for distinguishing OW from sea ice. Then, samples of sea ice, including FYI and MYI, are randomly split into training and test dataset. The features of the training set are used as input variables to train the FYI-MYI classifiers, which achieve an overall accuracy of 84.82% and 71.71% respectively by RF and SVM classifiers. Finally, the features in every month are used as training and testing set in turn to cross-validate the performance of the proposed classifier. The results indicate the strong sensitivity of GNSS signals to sea ice types and the great potential of ML classifiers for GNSS-R applications.

A Comparison of Machine Learning Algorithms for the Segmentation and Classification of Snow Micro Penetrometer Profiles on Arctic Sea Ice

2021 ◽  
Author(s):  
Julia Kaltenborn ◽  
Viviane Clay ◽  
Amy R. Macfarlane ◽  
Joshua Michael Lloyd King ◽  
Martin Schneebeli
Keyword(s):  
Machine Learning ◽  
Sea Ice ◽  
Arctic Sea Ice ◽  
Machine Learning Algorithms ◽  
Training Data ◽  
Support Vector ◽  
Snow Layer ◽  
Arctic Sea ◽  
Execution Speed

<p>Snow-layer classification is an essential diagnostic task for a wide variety of cryospheric science and climate research applications. Traditionally, these measurements are made in snow pits, requiring trained operators and a substantial time commitment. The SnowMicroPen (SMP), a portable high-resolution snow penetrometer, has been demonstrated as a capable tool for rapid snow grain classification and layer type segmentation through statistical inversion of its mechanical signal. The manual classification of the SMP profiles requires time and training and becomes infeasible for large datasets.</p><p>Here, we introduce a novel set of SMP measurements collected during the MOSAiC expedition and apply Machine Learning (ML) algorithms to automatically classify and segment SMP profiles of snow on Arctic sea ice. To this end, different supervised and unsupervised ML methods, including Random Forests, Support Vector Machines, Artificial Neural Networks, and k-means Clustering, are compared. A subsequent segmentation of the classified data results in distinct layers and snow grain markers for the SMP profiles. The models are trained with the dataset by King et al. (2020) and the MOSAiC SMP dataset. The MOSAiC dataset is a unique and extensive dataset characterizing seasonal and spatial variation of snow on the central Arctic sea-ice.</p><p>We will test and compare the different algorithms and evaluate the algorithms’ effectiveness based on the need for initial dataset labeling, execution speed, and ease of implementation. In particular, we will compare supervised to unsupervised methods, which are distinguished by their need for labeled training data.</p><p>The implementation of different ML algorithms for SMP profile classification could provide a fast and automatic grain type classification and snow layer segmentation. Based on the gained knowledge from the algorithms’ comparison, a tool can be built to provide scientists from different fields with an immediate SMP profile classification and segmentation. </p><p> </p><p>King, J., Howell, S., Brady, M., Toose, P., Derksen, C., Haas, C., & Beckers, J. (2020). Local-scale variability of snow density on Arctic sea ice. <em>The Cryosphere</em>, <em>14</em>(12), 4323-4339, https://doi.org/10.5194/tc-14-4323-2020.</p>

scholarly journals Extraction of Sea Ice Cover by Sentinel-1 SAR Based on SVM with Unsupervised Generation of Training Data

2020 ◽  
Author(s):  
Xiaoming Li ◽  
Yan Sun ◽  
Qiang Zhang
Keyword(s):  
Machine Learning ◽  
Sea Ice ◽  
Learning Algorithm ◽  
Texture Features ◽  
Open Water ◽  
Ice Cover ◽  
Training Data ◽  
Support Vector ◽  
Training Samples

In this paper, we focus on developing a novel method to extract sea ice cover (i.e., discrimination/classification of sea ice and open water) using Sentinel-1 (S1) cross-polarization (vertical-horizontal, VH or horizontal-vertical, HV) data in extra wide (EW) swath mode based on the machine learning algorithm support vector machine (SVM). The classification basis includes the S1 radar backscatter coefficients and texture features that are calculated from S1 data using the gray level co-occurrence matrix (GLCM). Different from previous methods where appropriate samples are manually selected to train the SVM to classify sea ice and open water, we proposed a method of unsupervised generation of the training samples based on two GLCM texture features, i.e. entropy and homogeneity, that have contrasting characteristics on sea ice and open water. We eliminate the most uncertainty of selecting training samples in machine learning and achieve automatic classification of sea ice and open water by using S1 EW data. The comparison shows good agreement between the SAR-derived sea ice cover using the proposed method and a visual inspection, of which the accuracy reaches approximately 90% - 95% based on a few cases. Besides this, compared with the analyzed sea ice cover data Ice Mapping System (IMS) based on 728 S1 EW images, the accuracy of extracted sea ice cover by using S1 data is more than 80%.

scholarly journals Regional melt-pond fraction and albedo of thin Arctic first-year drift ice in late summer

2015 ◽  
Vol 9 (1) ◽  
pp. 255-268 ◽  
Author(s):  
D. V. Divine ◽  
M. A. Granskog ◽  
S. R. Hudson ◽  
C. A. Pedersen ◽  
T. I. Karlsen ◽  
...  
Keyword(s):  
Sea Ice ◽  
Open Water ◽  
Ice Cover ◽  
Arctic Sea Ice ◽  
The Arctic ◽  
First Year ◽  
Surface Type ◽  
Melt Pond ◽  
Melt Ponds

Abstract. The paper presents a case study of the regional (≈150 km) morphological and optical properties of a relatively thin, 70–90 cm modal thickness, first-year Arctic sea ice pack in an advanced stage of melt. The study combines in situ broadband albedo measurements representative of the four main surface types (bare ice, dark melt ponds, bright melt ponds and open water) and images acquired by a helicopter-borne camera system during ice-survey flights. The data were collected during the 8-day ICE12 drift experiment carried out by the Norwegian Polar Institute in the Arctic, north of Svalbard at 82.3° N, from 26 July to 3 August 2012. A set of > 10 000 classified images covering about 28 km2 revealed a homogeneous melt across the study area with melt-pond coverage of ≈ 0.29 and open-water fraction of ≈ 0.11. A decrease in pond fractions observed in the 30 km marginal ice zone (MIZ) occurred in parallel with an increase in open-water coverage. The moving block bootstrap technique applied to sequences of classified sea-ice images and albedo of the four surface types yielded a regional albedo estimate of 0.37 (0.35; 0.40) and regional sea-ice albedo of 0.44 (0.42; 0.46). Random sampling from the set of classified images allowed assessment of the aggregate scale of at least 0.7 km2 for the study area. For the current setup configuration it implies a minimum set of 300 images to process in order to gain adequate statistics on the state of the ice cover. Variance analysis also emphasized the importance of longer series of in situ albedo measurements conducted for each surface type when performing regional upscaling. The uncertainty in the mean estimates of surface type albedo from in situ measurements contributed up to 95% of the variance of the estimated regional albedo, with the remaining variance resulting from the spatial inhomogeneity of sea-ice cover.

The Thinning of Arctic Sea Ice, 1988–2003: Have We Passed a Tipping Point?

2005 ◽  
Vol 18 (22) ◽  
pp. 4879-4894 ◽  
Author(s):  
R. W. Lindsay ◽  
J. Zhang
Keyword(s):  
Sea Ice ◽  
Open Water ◽  
Ocean Model ◽  
Tipping Point ◽  
Arctic Sea Ice ◽  
The Arctic ◽  
First Year ◽  
Positive Feedbacks ◽  
Cooling Period ◽  
Arctic Sea

Abstract Recent observations of summer Arctic sea ice over the satellite era show that record or near-record lows for the ice extent occurred in the years 2002–05. To determine the physical processes contributing to these changes in the Arctic pack ice, model results from a regional coupled ice–ocean model have been analyzed. Since 1988 the thickness of the simulated basinwide ice thinned by 1.31 m or 43%. The thinning is greatest along the coast in the sector from the Chukchi Sea to the Beaufort Sea to Greenland. It is hypothesized that the thinning since 1988 is due to preconditioning, a trigger, and positive feedbacks: 1) the fall, winter, and spring air temperatures over the Arctic Ocean have gradually increased over the last 50 yr, leading to reduced thickness of first-year ice at the start of summer; 2) a temporary shift, starting in 1989, of two principal climate indexes (the Arctic Oscillation and Pacific Decadal Oscillation) caused a flushing of some of the older, thicker ice out of the basin and an increase in the summer open water extent; and 3) the increasing amounts of summer open water allow for increasing absorption of solar radiation, which melts the ice, warms the water, and promotes creation of thinner first-year ice, ice that often entirely melts by the end of the subsequent summer. Internal thermodynamic changes related to the positive ice–albedo feedback, not external forcing, dominate the thinning processes over the last 16 yr. This feedback continues to drive the thinning after the climate indexes return to near-normal conditions in the late 1990s. The late 1980s and early 1990s could be considered a tipping point during which the ice–ocean system began to enter a new era of thinning ice and increasing summer open water because of positive feedbacks. It remains to be seen if this era will persist or if a sustained cooling period can reverse the processes.

scholarly journals Seasonal cycle and long-term trend of solar energy fluxes through Arctic sea ice

2014 ◽  
Vol 8 (6) ◽  
pp. 2219-2233 ◽  
Author(s):  
S. Arndt ◽  
M. Nicolaus
Keyword(s):  
Sea Ice ◽  
Open Water ◽  
Ice Cover ◽  
Reanalysis Data ◽  
Arctic Sea Ice ◽  
Light Transmittance ◽  
First Year ◽  
Sea Ice Concentration ◽  
Solar Heat ◽  
Arctic Sea

Abstract. Arctic sea ice has not only decreased in volume during the last decades, but has also changed in its physical properties towards a thinner and more seasonal ice cover. These changes strongly impact the energy budget, and might affect the ice-associated ecosystems. In this study, we quantify solar shortwave fluxes through sea ice for the entire Arctic during all seasons. To focus on sea-ice-related processes, we exclude fluxes through open water, scaling linearly with sea ice concentration. We present a new parameterization of light transmittance through sea ice for all seasons as a function of variable sea ice properties. The maximum monthly mean solar heat flux under the ice of 30 × 105 Jm−2 occurs in June, enough heat to melt 0.3 m of sea ice. Furthermore, our results suggest that 96% of the annual solar heat input through sea ice occurs during only a 4-month period from May to August. Applying the new parameterization to remote sensing and reanalysis data from 1979 to 2011, we find an increase in transmitted light of 1.5% yr−1 for all regions. This corresponds to an increase in potential sea ice bottom melt of 63% over the 33-year study period. Sensitivity studies reveal that the results depend strongly on the timing of melt onset and the correct classification of ice types. Assuming 2 weeks earlier melt onset, the annual transmitted solar radiation to the upper ocean increases by 20%. Continuing the observed transition from a mixed multi-year/first-year sea ice cover to a seasonal ice cover results in an increase in light transmittance by an additional 18%.

Bacterial Production in the Bottom Surface of Sea Ice in the Canadian Subarctic

1990 ◽  
Vol 47 (10) ◽  
pp. 1986-1995 ◽  
Author(s):  
J. N. Bunch ◽  
R. C. Harland
Keyword(s):  
Organic Carbon ◽  
Sea Ice ◽  
Open Water ◽  
Arctic Sea Ice ◽  
Trophic Levels ◽  
Bottom Surface ◽  
First Year ◽  
Bacterial Assemblage ◽  
Ice Surface ◽  
Arctic Sea

Standing stocks of bacteria in the bottom of first-year sea ice at Frobisher Bay, N.W.T., increased fivefold between March and May (1985 and 1986) and constituted up to 5% of particulate organic carbon (POC). Autoradiography demonstrated that approximately one-third of the bacterial assemblage incorporated radioactive thymidine. The mean volume of cells was six times larger than that in the underlying water, and the assemblage was dominated by rod-shaped cells rather than the coccus-shaped cells prevalent in the water column. Bacterial carbon production by 3H-thymidine incorporation amounted to 0.04 mg carbon m−2∙h−1, or a doubling time of about 22 h, in the bottom ice surface and 0.01 mg carbon m−3∙h−1 in the underlying water. The concentration of dissolved organic carbon (DOC) was generally much higher in the bottom ice surface than in the underlying water, and was closely related to rate of cell production. A model of bacterial dependancy on DOC derived from primary production suggests that bacteria are important in the localized production of POC in the bottom of arctic sea ice, and contribute to an early source of nutrition for higher trophic levels before summer production in open water.

scholarly journals Regional albedo of Arctic first-year drift ice in advanced stages of melt from the combination of in situ measurements and aerial imagery

2014 ◽  
Vol 8 (4) ◽  
pp. 3699-3732
Author(s):  
D. V. Divine ◽  
M. A. Granskog ◽  
S. R. Hudson ◽  
C. A. Pedersen ◽  
T. I. Karlsen ◽  
...  
Keyword(s):  
Sea Ice ◽  
Open Water ◽  
In Situ Measurements ◽  
Arctic Sea Ice ◽  
Aerial Imagery ◽  
The Arctic ◽  
First Year ◽  
Melt Ponds ◽  
Advanced Stages

Abstract. The paper presents a case study of the regional (≈ 150 km) broadband albedo of first year Arctic sea ice in advanced stages of melt, estimated from a combination of in situ albedo measurements and aerial imagery. The data were collected during the eight day ICE12 drift experiment carried out by the Norwegian Polar Institute in the Arctic north of Svalbard at 82.3° N from 26 July to 3 August 2012. The study uses in situ albedo measurements representative of the four main surface types: bare ice, dark melt ponds, bright melt ponds and open water. Images acquired by a helicopter borne camera system during ice survey flights covered about 28 km2. A subset of > 8000 images from the area of homogeneous melt with open water fraction of ≈ 0.11 and melt pond coverage of ≈ 0.25 used in the upscaling yielded a regional albedo estimate of 0.40 (0.38; 0.42). The 95% confidence interval on the estimate was derived using the moving block bootstrap approach applied to sequences of classified sea ice images and albedo of the four surface types treated as random variables. Uncertainty in the mean estimates of surface type albedo from in situ measurements contributed some 95% of the variance of the estimated regional albedo, with the remaining variance resulting from the spatial inhomogeneity of sea ice cover. The results of the study are of relevance for the modeling of sea ice processes in climate simulations. It particularly concerns the period of summer melt, when the optical properties of sea ice undergo substantial changes, which existing sea ice models have significant diffuculty accurately reproducing.

Arctic sea ice volume budget decomposition satellite product for the CryoSat-2 (2010-2020) period 

2021 ◽  
Author(s):  
Harry Heorton ◽  
Michel Tsamados ◽  
Paul Holland ◽  
Jack Landy
Keyword(s):  
Sea Ice ◽  
Climate Model ◽  
Volume Growth ◽  
Arctic Sea Ice ◽  
First Year ◽  
Sea Ice Concentration ◽  
Marginal Seas ◽  
Ice Volume ◽  
Arctic Sea ◽  
Ice Concentration

<p><span>We combine satellite-derived observations of sea ice concentration, drift, and thickness to provide the first observational decomposition of the dynamic (advection/divergence) and thermodynamic (melt/growth) drivers of wintertime Arctic sea ice volume change. Ten winter growth seasons are analyzed over the CryoSat-2 period between October 2010 and April 2020. Sensitivity to several observational products is performed to provide an estimated uncertainty of the budget calculations. The total thermodynamic ice volume growth and dynamic ice losses are calculated with marked seasonal, inter-annual and regional variations</span><span>. Ice growth is fastest during Autumn, in the Marginal Seas and over first year ice</span><span>. Our budget decomposition methodology can help diagnose the processes confounding climate model predictions of sea ice. We make our product and code available to the community in monthly pan-Arctic netcdft files for the entire October 2010 to April 2020 period.</span></p>

scholarly journals Albedo of Young and First-Year Antarctic Sea Ice

1990 ◽  
Vol 14 ◽  
pp. 331 ◽  
Author(s):  
Richard Brandt ◽  
Ian Allison ◽  
Stephen Warren
Keyword(s):  
Sea Ice ◽  
Radiation Flux ◽  
Open Water ◽  
First Year ◽  
Spectral Measurements ◽  
Antarctic Sea Ice ◽  
Antarctic Expedition ◽  
Downward Radiation ◽  
Radiation Measurements ◽  
Snow Thickness

Reflection of solar radiation was studied in the seasonal sea-ice zone off East Antarctica on a cruise of the Australian Antarctic Expedition, October-December 1988. Spectral and total albedos were measured for grease ice, nilas, young grey ice, grey-white ice, snow-covered ice, and open water. Spectral measurements covered the region 400–1000 nm wavelength. For ice too thin to support our weight, the radiometers were mounted at the end of a 1.5 m rod extended out the door of a helicopter or from a basket hung from the ship's crane, using a positioning and leveling rack. Corrections had to be applied to the downward radiation flux because the helicopter or the crane was in the field of view of the cosine-collector. The fractional coverage of each of the ice types (and open water) was estimated hourly for the region near the ship, as well as the thickness of each ice type, and the snow thickness. Observations were carried out continuously during the four weeks the ship was in the ice, supplemented by occasional helicopter surveys covering larger areas. These observations, together with the radiation measurements, make possible the computation of area-average albedo for the East Antarctic sea-ice zone in spring.

Sign in / Sign up

Export Citation Format

Share Document