Debris-covered glaciers present significant challenges for accu-rately mapping and monitoring glacier dynamics,particularly in regions like the Eastern Pamir Plateau.This study shows a new hybrid ensemble classifier t...Debris-covered glaciers present significant challenges for accu-rately mapping and monitoring glacier dynamics,particularly in regions like the Eastern Pamir Plateau.This study shows a new hybrid ensemble classifier that uses random forest and decision tree algorithms to make mapping debris-covered glaciers more accurate using data from multiple satellites.The method leverages features derived from the SDGSAT-1,Sentinel-2,ASTER GDEM,and ITS_LIVE datasets,including color,texture,topography,land surface temperature,and velocity data.Conventional glacier mapping tech-niques often misclassify debris-covered areas due to their spectral similarity to the surrounding terrain,making this work crucial in addressing these limitations.To improve the accuracy of recogni-tion between debris-covered glaciers and non-glaciated areas by capitalizing on the strengths of multiple machine-learning algo-rithms and diverse data sources.The hybrid ensemble classifier did better than single-classifier models,with an overall accuracy of 92%and a Kappa coefficient of 0.885.It successfully delineated debris-covered glacier boundaries that closely matched established glacier inventories while offering a more detailed mapping of deb-ris extent.Key innovations in this research include integrating SDGSAT-1 data,which opens new avenues for glacier monitoring,and the development of an advanced feature selection strategy that enhances classification accuracy.Further,the study introduces new spectral indices and temperature-based metrics specifically designed for debris-covered glacier identification.This was a signifi cant step forward from previous work in the area.展开更多
The Antarctic Ice Sheet(AIS)has been losing ice mass and contributing to the rise in the global sea-level(GSL)for the last 4 decades,as quantified by using satellite observations.We developed a framework for implement...The Antarctic Ice Sheet(AIS)has been losing ice mass and contributing to the rise in the global sea-level(GSL)for the last 4 decades,as quantified by using satellite observations.We developed a framework for implementing the state-of-the-art input-output(IO)method that has the advantage of explicit estimation of the mass balance of individual glaciers,basins and the continent.We estimated the mass balance of the AIS from 2013 to 2018 using improved observations and updated datasets recently made available,including annual ice flow velocity maps from the Inter-mission Time Series of Land Ice Velocity and Elevation(ITS_LIVE)dataset,the Bed Machine and the Princess Elizabeth Land(PEL)Earth System Science Data(ESSD)datasets,and the surface mass balance from the RACMO 2.3 system.For example,using the improved ice thickness data,the proposed method for ice discharge estimation enables a 10%reduction of uncertainty in ice discharge.During the period of 2013–2018,an ice discharge acceleration of 6.9±6.5 Gt yr^(–2)in West Antarctica(WA)was detected,which contributed significantly to the estimated mass loss of~1069 Gt(–178.2±108.9 Gt yr^(–1))in the AIS.On the other hand,Queen Maud Land,East Antarctica(EA),showed clearly a mass gain rate of 56.0±10.0 Gt yr^(–1)due to the regional increase in surface mass balance.Our results extended the estimation period by 3 years in comparison to the published study using the same annual velocity maps from the ITS_LIVE dataset.Furthermore,our results,along with those from other studies using the IO method,reassures the acceleration of recent mass loss in WA and Wilkes Land in EA,which are caused by glacier thinning and ice shelf basal melting.Compared with the long-term mass balance record since 1979,our results suggest that the mass loss in AIS accelerated in the last decade.The developed framework can be modified for mass balance estimation of the AIS or for other ice sheets by using velocity maps from other satellite data or from different periods.展开更多
基金supported by the Joint Project of the Chinese Academy of Sciences(CAS)entitled Using Earth Observations to Address Ecology and Environment Change in the Pan-Antarctic Cryosphere[Grant No.183611KYSB20200059]the National Natural Science Foundation of China[Grant No.42276252].
文摘Debris-covered glaciers present significant challenges for accu-rately mapping and monitoring glacier dynamics,particularly in regions like the Eastern Pamir Plateau.This study shows a new hybrid ensemble classifier that uses random forest and decision tree algorithms to make mapping debris-covered glaciers more accurate using data from multiple satellites.The method leverages features derived from the SDGSAT-1,Sentinel-2,ASTER GDEM,and ITS_LIVE datasets,including color,texture,topography,land surface temperature,and velocity data.Conventional glacier mapping tech-niques often misclassify debris-covered areas due to their spectral similarity to the surrounding terrain,making this work crucial in addressing these limitations.To improve the accuracy of recogni-tion between debris-covered glaciers and non-glaciated areas by capitalizing on the strengths of multiple machine-learning algo-rithms and diverse data sources.The hybrid ensemble classifier did better than single-classifier models,with an overall accuracy of 92%and a Kappa coefficient of 0.885.It successfully delineated debris-covered glacier boundaries that closely matched established glacier inventories while offering a more detailed mapping of deb-ris extent.Key innovations in this research include integrating SDGSAT-1 data,which opens new avenues for glacier monitoring,and the development of an advanced feature selection strategy that enhances classification accuracy.Further,the study introduces new spectral indices and temperature-based metrics specifically designed for debris-covered glacier identification.This was a signifi cant step forward from previous work in the area.
基金supported by the National Key Research&Development Program of China(Grant No.2017YFA0603102)the National Natural Science Foundation of China(Grant Nos.41730102,41771471,41941006&4201101408)+1 种基金the Shanghai Science and Technology Development Funds(Grant No.21ZR1469700)supported by the Central University Research Fund。
文摘The Antarctic Ice Sheet(AIS)has been losing ice mass and contributing to the rise in the global sea-level(GSL)for the last 4 decades,as quantified by using satellite observations.We developed a framework for implementing the state-of-the-art input-output(IO)method that has the advantage of explicit estimation of the mass balance of individual glaciers,basins and the continent.We estimated the mass balance of the AIS from 2013 to 2018 using improved observations and updated datasets recently made available,including annual ice flow velocity maps from the Inter-mission Time Series of Land Ice Velocity and Elevation(ITS_LIVE)dataset,the Bed Machine and the Princess Elizabeth Land(PEL)Earth System Science Data(ESSD)datasets,and the surface mass balance from the RACMO 2.3 system.For example,using the improved ice thickness data,the proposed method for ice discharge estimation enables a 10%reduction of uncertainty in ice discharge.During the period of 2013–2018,an ice discharge acceleration of 6.9±6.5 Gt yr^(–2)in West Antarctica(WA)was detected,which contributed significantly to the estimated mass loss of~1069 Gt(–178.2±108.9 Gt yr^(–1))in the AIS.On the other hand,Queen Maud Land,East Antarctica(EA),showed clearly a mass gain rate of 56.0±10.0 Gt yr^(–1)due to the regional increase in surface mass balance.Our results extended the estimation period by 3 years in comparison to the published study using the same annual velocity maps from the ITS_LIVE dataset.Furthermore,our results,along with those from other studies using the IO method,reassures the acceleration of recent mass loss in WA and Wilkes Land in EA,which are caused by glacier thinning and ice shelf basal melting.Compared with the long-term mass balance record since 1979,our results suggest that the mass loss in AIS accelerated in the last decade.The developed framework can be modified for mass balance estimation of the AIS or for other ice sheets by using velocity maps from other satellite data or from different periods.
