![]() Therefore, this study tests a new automatic classification scheme for hierarchical mapping of glacier surfaces based on machine learning classifiers including k-nearest neighbors (KNN), support vector machine (SVM), gradient boosting (GB), decision tree (DT), random forest (RF) and multi-layer perceptron (MLP). These findings confirmed a relatively stable behavior of glaciers in this area since the LIA and a strong control of local geomorphometric characteristics on glacial change.ĭebris cover on glacier surfaces hampers the accurate detection of debris-covered ice using traditional techniques based on image band ratios. Compared to other areas on the Tibetan Plateau and its surrounding mountains, glacial changes have been less extensive in the Hunza Basin since the LIA. In contrast, the change in ELA was not strongly affected by glacial area but had some correlations with slope and mean elevation. The relative area reduction was strongly affected by geomorphometric factors and small glaciers tended to experience relatively large areal reductions. Our results revealed 6.6% decrease in the total area, 14.5% shortening in the total length along the flowline of each LIA glacier, and approximately 98 m rising in the mean ELA since the LIA. ![]() We also derived geomorphometric parameters for each glacier using the Shuttle Radar Topographic Mission digital elevation model and reconstructed the equilibrium line altitudes (ELAs) for both contemporary and LIA glaciers based on the toe-to-ridge altitude method. We mapped 432 contemporary glaciers and their corresponding 313 LIA glaciers in the Hunza Basin, western Karakoram, using Google Earth and Landsat images. This study investigated glacial changes in the Karakoram since the Little Ice Age (LIA) to examine if this relatively stable scenario has occurred for a longer term. Glaciers have remained relatively stable in the Karakoram over the past few decades, in contrast to rapid retreat in other regions of the Tibetan Plateau and its surrounding mountains. Therefore, manual editing was necessary to improve the final glacier maps. Furthermore, final glacier maps show satisfactory mapping results, but identification of the debris-cover glacier terminus (covered by thick debris layer) is still problematic. Glacier areas derived using the proposed approach were 3% less than in the reference datasets. Accuracy of the generated glacier outlines was assessed through comparison with glacier outlines based on the Second Chinese Glacier Inventory (SCGI) data and glacier outlines created from high-resolution Google Earth™ images of 2009. Final vector maps of the glaciers were created using overlay tools in a geographic information system (GIS). The debris-free glaciers were identified using the band ratio (TM band 4/TM band 5) technique. For accurate mapping of supraglacial debris area, clustering results were combined with a thermal mask generated from the Landsat TM thermal band. Then they were organized in similar surface groups using cluster analysis. ![]() The geomorphometric parameters slope, plan, and profile curvature were generated from ASTER GDEM. For this study, we applied a multi-criteria technique to map the glaciers of the Shaksgam valley of China, using Landsat Thematic Mapper (Landsat TM) (2009) and Advanced Spaceborne Thermal Emission and Reflection Radiometer Global Digital Elevation Model version two (ASTER GDEM V2) data. Repeated creation of the glacier inventories is important to assess glacier–climate interactions and to predict future runoff from glacierized catchments. Glaciers in the Shaksgam valley provide important fresh water resources to neighborhood livelihood.
0 Comments
Leave a Reply. |