scholarly journals Applying high-resolution visible imagery to satellite melt pond fraction retrieval: A neural network approach

2018 ◽  
Vol 3 (3) ◽  
Author(s):  
Qi Liu 1 ◽  
Yawen Zhang 1
Keyword(s):  
Neural Network ◽  
Sea Ice ◽  
Satellite Data ◽  
Arctic Sea Ice ◽  
The Arctic ◽  
Neural Network Approach ◽  
Melt Pond ◽  
Melt Ponds ◽  
Arctic Sea ◽  
Visible Imagery

During summer, melt ponds have a significant influence on Arctic sea-ice albedo. The melt pond fraction (MPF) also has the ability to forecast the Arctic sea-ice in a certain period. It is important to retrieve accurate melt pond fraction (MPF) from satellite data for Arctic research. This paper proposes a satellite MPF retrieval model based on the multi-layer neural network, named MPF-NN. Our model uses multi-spectral satellite data as model input and MPF information from multi-site and multi-period visible imagery as prior knowledge for modeling. It can effectively model melt ponds evolution of different regions and periods over the Arctic. Evaluation results show that the MPF retrieved from MODIS data using the proposed model has an RMSE of 3.91% and a correlation coefficient of 0.73. The seasonal distribution of MPF is also consistent with previous results.

scholarly journals Inorganic carbon dynamics of melt pond-covered first year sea ice in the Canadian Arctic

2014 ◽  
Vol 11 (5) ◽  
pp. 7485-7519 ◽  
Author(s):  
N.-X. Geilfus ◽  
R. J. Galley ◽  
O. Crabeck ◽  
T. Papakyriakou ◽  
J. Landy ◽  
...  
Keyword(s):  
Sea Ice ◽  
Arctic Sea Ice ◽  
The Arctic ◽  
First Year ◽  
Ice Melt ◽  
Melt Pond ◽  
Melt Ponds ◽  
Arctic Sea ◽  
Melt Water

Abstract. Melt pond formation is a common feature of the spring and summer Arctic sea ice. However, the role of the melt ponds formation and the impact of the sea ice melt on both the direction and size of CO2 flux between air and sea is still unknown. Here we describe the CO2-carbonate chemistry of melting sea ice, melt ponds and the underlying seawater associated with measurement of CO2 fluxes across first year landfast sea ice in the Resolute Passage, Nunavut, in June 2012. Early in the melt season, the increase of the ice temperature and the subsequent decrease of the bulk ice salinity promote a strong decrease of the total alkalinity (TA), total dissolved inorganic carbon (TCO2) and partial pressure of CO2 (pCO2) within the bulk sea ice and the brine. Later on, melt pond formation affects both the bulk sea ice and the brine system. As melt ponds are formed from melted snow the in situ melt pond pCO2 is low (36 μatm). The percolation of this low pCO2 melt water into the sea ice matrix dilutes the brine resulting in a strong decrease of the in situ brine pCO2 (to 20 μatm). As melt ponds reach equilibrium with the atmosphere, their in situ pCO2 increase (up to 380 μatm) and the percolation of this high concentration pCO2 melt water increase the in situ brine pCO2 within the sea ice matrix. The low in situ pCO2 observed in brine and melt ponds results in CO2 fluxes of −0.04 to −5.4 mmol m–2 d–1. As melt ponds reach equilibrium with the atmosphere, the uptake becomes less significant. However, since melt ponds are continuously supplied by melt water their in situ pCO2 still remains low, promoting a continuous but moderate uptake of CO2 (~ −1mmol m–2 d–1). The potential uptake of atmospheric CO2 by melting sea ice during the Arctic summer has been estimated from 7 to 16 Tg of C ignoring the role of melt ponds. This additional uptake of CO2 associated to Arctic sea ice needs to be further explored and considered in the estimation of the Arctic Ocean's overall CO2 budget.

scholarly journals Inorganic carbon dynamics of melt-pond-covered first-year sea ice in the Canadian Arctic

2015 ◽  
Vol 12 (6) ◽  
pp. 2047-2061 ◽  
Author(s):  
N.-X. Geilfus ◽  
R. J. Galley ◽  
O. Crabeck ◽  
T. Papakyriakou ◽  
J. Landy ◽  
...  
Keyword(s):  
Sea Ice ◽  
Arctic Sea Ice ◽  
The Arctic ◽  
First Year ◽  
Co2 Fluxes ◽  
Ice Melt ◽  
Melt Pond ◽  
Melt Ponds ◽  
Arctic Sea

Abstract. Melt pond formation is a common feature of spring and summer Arctic sea ice, but the role and impact of sea ice melt and pond formation on both the direction and size of CO2 fluxes between air and sea is still unknown. Here we report on the CO2–carbonate chemistry of melting sea ice, melt ponds and the underlying seawater as well as CO2 fluxes at the surface of first-year landfast sea ice in the Resolute Passage, Nunavut, in June 2012. Early in the melt season, the increase in ice temperature and the subsequent decrease in bulk ice salinity promote a strong decrease of the total alkalinity (TA), total dissolved inorganic carbon (TCO2) and partial pressure of CO2 (pCO2) within the bulk sea ice and the brine. As sea ice melt progresses, melt ponds form, mainly from melted snow, leading to a low in situ melt pond pCO2 (36 μatm). The percolation of this low salinity and low pCO2 meltwater into the sea ice matrix decreased the brine salinity, TA and TCO2, and lowered the in situ brine pCO2 (to 20 μatm). This initial low in situ pCO2 observed in brine and melt ponds results in air–ice CO2 fluxes ranging between −0.04 and −5.4 mmol m−2 day−1 (negative sign for fluxes from the atmosphere into the ocean). As melt ponds strive to reach pCO2 equilibrium with the atmosphere, their in situ pCO2 increases (up to 380 μatm) with time and the percolation of this relatively high concentration pCO2 meltwater increases the in situ brine pCO2 within the sea ice matrix as the melt season progresses. As the melt pond pCO2 increases, the uptake of atmospheric CO2 becomes less significant. However, since melt ponds are continuously supplied by meltwater, their in situ pCO2 remains undersaturated with respect to the atmosphere, promoting a continuous but moderate uptake of CO2 (~ −1 mmol m−2 day−1) into the ocean. Considering the Arctic seasonal sea ice extent during the melt period (90 days), we estimate an uptake of atmospheric CO2 of −10.4 Tg of C yr−1. This represents an additional uptake of CO2 associated with Arctic sea ice that needs to be further explored and considered in the estimation of the Arctic Ocean's overall CO2 budget.

scholarly journals Retrieval of Melt Pond Fraction over Arctic Sea Ice during 2000–2019 Using an Ensemble-Based Deep Neural Network

2020 ◽  
Vol 12 (17) ◽  
pp. 2746
Author(s):  
Yifan Ding ◽  
Xiao Cheng ◽  
Jiping Liu ◽  
Fengming Hui ◽  
Zhenzhan Wang ◽  
...  
Keyword(s):  
Neural Network ◽  
Sea Ice ◽  
Deep Neural Network ◽  
Arctic Sea Ice ◽  
The Arctic ◽  
Sea Ice Concentration ◽  
Melt Pond ◽  
Arctic Sea ◽  
Target Data ◽  
Melting Season

The accurate knowledge of variations of melt ponds is important for understanding the Arctic energy budget due to its albedo–transmittance–melt feedback. In this study, we develop and validate a new method for retrieving melt pond fraction (MPF) over Arctic sea ice using all seven spectral bands of MODIS surface reflectance. We construct a robust ensemble-based deep neural network and use in-situ MPF observations collected from multiple sources as the target data to train the network. We examine the potential influence of using sea ice concentration (SIC) from different sources as additional target data (besides MPF) on the MPF retrieval. The results suggest that the inclusion of SIC has a minor impact on MPF retrieval. Based on this, we create a new MPF data from 2000 to 2019 (the longest data in our knowledge). The validation shows that our new MPF data is in good agreement with the observations. We further compare the new MPF dataset with the previously published MPF datasets. It is found that the evolution of the new MPF is similar to previous MPF data throughout the melting season, but the new MPF data is in relatively better agreement with the observations in terms of correlations and root mean squared errors (RMSE), and also has the smallest value in the first half of the melting season.

scholarly journals Application of Aerosondes to Melt-Pond Observations over Arctic Sea Ice

2008 ◽  
Vol 25 (2) ◽  
pp. 327-334 ◽  
Author(s):  
Jun Inoue ◽  
Judith A. Curry ◽  
James A. Maslanik
Keyword(s):  
Sea Ice ◽  
Arctic Sea Ice ◽  
The Arctic ◽  
First Year ◽  
Melt Pond ◽  
Earth Observing System ◽  
Melt Ponds ◽  
Arctic Sea ◽  
Ice Concentration ◽  
Aerial Vehicle

Abstract Continuous observation of sea ice using a small robotic aircraft called the Aerosonde was made over the Arctic Ocean from Barrow, Alaska, on 20–21 July 2003. Over a region located 350 km off the coast of Barrow, images obtained from the aircraft were used to characterize the sea ice and to determine the fraction of melt ponds on both multiyear and first-year ice. Analysis of the data indicates that melt-pond fraction increased northward from 20% to 30% as the ice fraction increased. However, the fraction of ponded ice was over 30% in the multiyear ice zone while it was about 25% in the first-year ice zone. A comparison with a satellite microwave product showed that the ice concentration derived from the Advanced Microwave Scanning Radiometer for the Earth Observing System (AMSR-E) has a negative bias of 7% due to melt ponds. These analyses demonstrate the utility of recent advances in unmanned aerial vehicle (UAV) technology for monitoring and interpreting the spatial variations in the sea ice with melt ponds.

scholarly journals Retrieval of Snow Depth over Arctic Sea Ice Using a Deep Neural Network

2019 ◽  
Vol 11 (23) ◽  
pp. 2864 ◽  
Author(s):  
Jiping Liu ◽  
Yuanyuan Zhang ◽  
Xiao Cheng ◽  
Yongyun Hu
Keyword(s):  
Neural Network ◽  
Sea Ice ◽  
Snow Depth ◽  
Deep Neural Network ◽  
Arctic Basin ◽  
Temporal Variations ◽  
Arctic Sea Ice ◽  
The Arctic ◽  
Sea Ice Thickness ◽  
Arctic Sea

The accurate knowledge of spatial and temporal variations of snow depth over sea ice in the Arctic basin is important for understanding the Arctic energy budget and retrieving sea ice thickness from satellite altimetry. In this study, we develop and validate a new method for retrieving snow depth over Arctic sea ice from brightness temperatures at different frequencies measured by passive microwave radiometers. We construct an ensemble-based deep neural network and use snow depth measured by sea ice mass balance buoys to train the network. First, the accuracy of the retrieved snow depth is validated with observations. The results show the derived snow depth is in good agreement with the observations, in terms of correlation, bias, root mean square error, and probability distribution. Our ensemble-based deep neural network can be used to extend the snow depth retrieval from first-year sea ice (FYI) to multi-year sea ice (MYI), as well as during the melting period. Second, the consistency and discrepancy of snow depth in the Arctic basin between our retrieval using the ensemble-based deep neural network and two other available retrievals using the empirical regression are examined. The results suggest that our snow depth retrieval outperforms these data sets.

scholarly journals Deep Learning Convolutional Neural Network Applying for the Arctic Acoustic Tomography Current Inversion Accuracy Improvement

2021 ◽  
Vol 9 (7) ◽  
pp. 755
Author(s):  
Kangkang Jin ◽  
Jian Xu ◽  
Zichen Wang ◽  
Can Lu ◽  
Long Fan ◽  
...  
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Sea Ice ◽  
Convolutional Neural Network ◽  
Arctic Sea Ice ◽  
The Arctic ◽  
Small Scale ◽  
Acoustic Tomography ◽  
Arctic Sea ◽  
Current Inversion

Warm current has a strong impact on the melting of sea ice, so clarifying the current features plays a very important role in the Arctic sea ice coverage forecasting study field. Currently, Arctic acoustic tomography is the only feasible method for the large-range current measurement under the Arctic sea ice. Furthermore, affected by the high latitudes Coriolis force, small-scale variability greatly affects the accuracy of Arctic acoustic tomography. However, small-scale variability could not be measured by empirical parameters and resolved by Regularized Least Squares (RLS) in the inverse problem of Arctic acoustic tomography. In this paper, the convolutional neural network (CNN) is proposed to enhance the prediction accuracy in the Arctic, and especially, Gaussian noise is added to reflect the disturbance of the Arctic environment. First, we use the finite element method to build the background ocean model. Then, the deep learning CNN method constructs the non-linear mapping relationship between the acoustic data and the corresponding flow velocity. Finally, the simulation result shows that the deep learning convolutional neural network method being applied to Arctic acoustic tomography could achieve 45.87% accurate improvement than the common RLS method in the current inversion.

scholarly journals Mapping the Bathymetry of Melt Ponds on Arctic Sea Ice Using Hyperspectral Imagery

2020 ◽  
Vol 12 (16) ◽  
pp. 2623 ◽  
Author(s):  
Marcel König ◽  
Gerit Birnbaum ◽  
Natascha Oppelt
Keyword(s):  
Sea Ice ◽  
Hyperspectral Imagery ◽  
Atmospheric Correction ◽  
Hyperspectral Data ◽  
Arctic Sea Ice ◽  
Atmospheric Conditions ◽  
Melt Pond ◽  
Melt Ponds ◽  
Arctic Sea ◽  
Highly Correlated

Hyperspectral remote-sensing instruments on unmanned aerial vehicles (UAVs), aircraft and satellites offer new opportunities for sea ice observations. We present the first study using airborne hyperspectral imagery of Arctic sea ice and evaluate two atmospheric correction approaches (ATCOR-4 (Atmospheric and Topographic Correction version 4; v7.0.0) and empirical line calibration). We apply an existing, field data-based model to derive the depth of melt ponds, to airborne hyperspectral AisaEAGLE imagery and validate results with in situ measurements. ATCOR-4 results roughly match the shape of field spectra but overestimate reflectance resulting in high root-mean-square error (RMSE) (between 0.08 and 0.16). Noisy reflectance spectra may be attributed to the low flight altitude of 200 ft and Arctic atmospheric conditions. Empirical line calibration resulted in smooth, accurate spectra (RMSE < 0.05) that enabled the assessment of melt pond bathymetry. Measured and modeled pond bathymetry are highly correlated (r = 0.86) and accurate (RMSE = 4.04 cm), and the model explains a large portion of the variability (R2 = 0.74). We conclude that an accurate assessment of melt pond bathymetry using airborne hyperspectral data is possible subject to accurate atmospheric correction. Furthermore, we see the necessity to improve existing approaches with Arctic-specific atmospheric profiles and aerosol models and/or by using multiple reference targets on the ground.

scholarly journals Comparison of different retrieval techniques for melt ponds on Arctic sea ice from Landsat and MODIS satellite data

2011 ◽  
Vol 52 (57) ◽  
pp. 185-191 ◽  
Author(s):  
Anja Rösel ◽  
Lars Kaleschke
Keyword(s):  
Sea Ice ◽  
Principal Component ◽  
Arctic Sea Ice ◽  
Additional Advantage ◽  
Melt Pond ◽  
Spectral Bands ◽  
Melt Ponds ◽  
Arctic Sea ◽  
Very High Spatial Resolution ◽  
Landsat 7

AbstractMelt ponds are regularly observed on the surface of Arctic sea ice in late spring and summer. They strongly reduce the surface albedo and accelerate the decay of Actic sea ice. Until now, only a few studies have looked at the spatial extent of melt ponds on Arctic sea ice. Knowledge of the melt-pond distribution on the entire Arctic sea ice would provide a solid basis for the parameterization of melt ponds in existing sea-ice models. Due to the different spectral properties of snow, ice and water, a multispectral sensor such as Landsat 7 ETM+ is generally applicable for the analysis of distribution. an additional advantage of the ETM+ sensor is the very high spatial resolution (30 m). an algorithm based on a principal component analysis (PCA) of two spectral channels has been developed in order to determine the melt-pond fraction. PCA allows differentiation of melt ponds and other surface types such as snow, ice or water. Spectral bands 1 and 4 with central wavelengths at 480 and 770 nm, respectively, are used as they represent the differences in the spectral albedo of melt ponds. A Landsat 7 ETM+ scene from 19 July 2001 was analysed using PCA. the melt-pond fraction determined by the PCA method yields a different spatial distribution of the ponded areas from that developed by others. A MODIS subset from the same date and area is also analysed. the classification of MODIS data results in a higher melt-pond fraction than both Landsat classifications.

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.

Sign in / Sign up

Export Citation Format

Share Document