DefinePK

Abstract

Glaciers in the Hindu Kush-Karakoram-Himalaya region impact Earth's climate, contribute freshwater downstream, and influence weather patterns of precipitation and temperature. However, the region needs more detailed information about its glaciers. Specifically, the stability of glaciers in the Karakoram range of the Hunza sub-basin is a well-known anomaly. Therefore, monitoring its glaciers is needed to understand the dynamics of climate change in HKH. Glacier inventory is baseline data for monitoring, and the clear-ice surface is a quantifying parameter of glacier changes. Recently, Operational Land Imager (OLI), exploited with machine learning (ML), is highly recommended for glacier monitoring due to improved accuracy. So, it is necessary to update the current status of glaciers in sub-basin using OLI and ML. Therefore, the study aims a) to evaluate the current extent of clear ice in the sub-basin to examine stability and b) to exploit the application of ML for extracting clear ice from OLI and assess accuracy. Google Earth environment is used to derive the data of Optical Land Imager and further analyze it with a machine learning approach to classify the extent of clear ice. Random Forest classifier with minimum Root means square error (0.1 to 0.4) used through SNAP environment. Results indicate satisfactory spatial distribution of clear ice in higher elevations (> 5000 meters). 10 % area difference percentage exhibited in overall extent; however, 28 glaciers (area > 5 km2) showed variation in the extent and confirmed the localized heterogeneity. Overall accuracy (82% to 83 %) and kappa coefficient values (0.64 to 0.65) confirm the role of individual bands of OLI. It is concluded that the glaciers in the sub-basin have an overall stable clear-ice extent except for variations in terminal ends. Meanwhile, machine learning has a significant role in the automatic extraction of clear ice when exploited with the OLI.