HR: 0800h
AN: GC41A-0088    [Abstracts]
TI: Integration of Classification Tree Analyses and Spatial Metrics to Assess Changes in Supraglacial Lakes in the Karakoram Himalaya
AU: * Bulley, H N
EM: hbulley@mail.unomaha.edu
AF: Department of Geography and Geology, University of Nebraska at Omaha, Omaha, NE 68182, United States
AU: Bishop, M P
EM: mpbishop@mail.unomaha.edu
AF: Department of Geography and Geology, University of Nebraska at Omaha, Omaha, NE 68182, United States
AU: Shroder, J F
EM: jshroder@mail.unomaha.edu
AF: Department of Geography and Geology, University of Nebraska at Omaha, Omaha, NE 68182, United States
AU: Haritashya, U K
EM: uharitashya@mail.unomaha.edu
AF: Department of Geography and Geology, University of Nebraska at Omaha, Omaha, NE 68182, United States
AB: Alpine glacier responses to climate chnage reveal increases in retreat with corresponding increases in production of glacier melt water and development of supraglacial lakes. The rate of occurrence and spatial extent of lakes in the Himalaya are difficult to determine because current spectral-based image analysis of glacier surfaces are limited through anisotropic reflectance and lack of high quality digital elevation models. Additionally, the limitations of multivariate classification algorithms to adequately segregate glacier features in satellite imagery have led to an increased interest in non-parametric methods, such as classification and regression trees. Our objectives are to demonstrate the utility of a semi-automated approach that integrates classification- tree-based image segmentation and object-oriented analysis to differentiate supraglacial lakes from glacier debris, ice cliffs, lateral and medial moraines. The classification-tree process involves a binary, recursive, partitioning non-parametric method that can account for non-linear relationships. We used 2002 and 2004 ASTER VNIR and SWIR imagery to assess the Baltoro Glacier in the Karakoram Himalaya. Other input variables include the normalized difference water index (NDWI), ratio images, Moran's I image, and fractal dimension. The classification tree was used to generate initial image segments and it was particularly effective in differentiating glacier features. The object-oriented analysis included the use of shape and spatial metrics to refine the classification-tree output. Classification-tree results show that NDWI is the most important single variable for characterizing the glacier-surface features, followed by NIR/IR ratio, IR band, and IR/Red ratio variables. Lake features extracted from both images show there were 142 lakes in 2002 as compared to 188 lakes in 2004. In general, there was a significant increase in planimetric area from 2002 to 2004, and we documented the formation of 46 new lakes. It appears that lake-size increments occur mostly in the lower part of the ablation zone, whereas most of the new lakes are formed in the upper part of the ablation zone. The classification-tree outputs are intuitive and the data-derived thresholds eliminate commonly subjective visual determination of threshold values. Semi-automated methods thus have the potential of eliminating laborious visual multi-temporal analysis of glacier-surface change, thereby producing consistent and replicable results needed to assess the trends of alpine-glacier response to climate change in the Himalaya.
DE: 0720 Glaciers
DE: 0758 Remote sensing
DE: 0762 Mass balance (1218, 1223)
DE: 0776 Glaciology (1621, 1827, 1863)
DE: 1621 Cryospheric change (0776)
SC: Global Environmental Change [GC]
MN: 2007 Fall Meeting