5:15 PM - 6:45 PM
[ACC26-P08] Forecasting Glacier Lake Outbursts Days in Advance Using Upstream Glacier Surface Temperature
Keywords:forecasting glacier lake outbursts, glacial surface temperature, satellite remote sensing
Glacier lake outburst floods (GLOFs) pose significant threats to downstream areas, leading to substantial damage to human lives and infrastructure. The high-altitude regions of Asia harbor numerous glacier lakes, elevating the risk of flooding. The Shisper Glacier Lake in northern Pakistan, specifically in the Hunza region, has experienced large-scale outbursts annually from 2019 to 2022, resulting in downstream river flow increases, bridge collapses, and extensive flooding hundreds of kilometers downstream. To mitigate such damages, early prediction of glacier lake outbursts, along with accurate alert systems, is crucial.
Previous studies have reported a risk assessment model utilizing Himalayan region satellite imagery to estimate glacier lake outbursts from moraine-dammed lakes with high precision. This model considers factors such as topography, watershed area, estimated water storage based on glacial lake area, and downstream river gradient. However, it is not specifically designed for issuing alerts for dam failures, and therefore lacks the temporal resolution necessary to provide timely alerts before a collapse occurs. Recently, another research group demonstrated cases where glacial lake outbursts occurred within a few days of a significant increase in surface temperature (approximately 0.3°C/day) in the upstream glacier. The mechanisms of collapse involve various factors and regional dependencies, including glacial lake overflow due to meteorological factors, outbursts, avalanches of ice or rocks into the lake, among others. Consequently, a comprehensive understanding of crucial meteorological elements and the temporal resolution of dam failure estimation models is not yet sufficient to issue appropriate alerts.
This study aims to establish a high-precision glacial lake dam failure estimation method that can predict the collapse of moraine-dammed glacial lakes several days in advance. As the initial step towards this goal, we focused on the glacial lake area, which is believed to have a strong correlation with the collapse of moraine-dammed glacial lakes, and examined its correlation with the surface temperature of the upstream glacier. The study area includes the Shisper Glacier Lake in Hunza, Pakistan, and the Muchuwar Glacier, considered to be the water source for this glacier lake. Due to the larger area of the upstream glacier compared to the glacial lake, estimating the surface temperature of the glacier can be crucial for assessing the risk of collapse caused by the expansion of the glacial lake, even in situations where satellite image analysis is hindered by cloud cover over the glacial lake. This information is deemed essential for issuing timely alerts.
Analysis utilizes Landsat 7 and 8 satellite images (a total of 11 images) captured approximately one month prior to the reported Shisper Glacier Lake outbursts in 2019, 2020, and 2022. Surface temperatures of the Muchuwar Glacier were calculated from Landsat 7 band 6 (10.40-12.50 µm) and Landsat 8 band 10 (10.6-11.19 µm) images. The Shisper Glacier Lake area was determined from Landsat 8 visible composite images.
From the upstream edge of the glacier lake, a correlation analysis was conducted between the surface temperature of glaciers within a range of 2.5 km intervals, spanning 2.5 to 12.5 km. The results revealed a high correlation, with r values ranging from 0.65 to 0.90. These findings indicate that surface temperature is a precise indicator for predicting glacier lake outbursts, and the study discusses the potential application of these results to glacier lakes in other regions.
This research was partially supported by the Telecommunications Advancement Foundation under the "Development of Human Resource Education Methods with SDG Problem-Solving Capabilities Using Both ICT and Hands-on Approach (Shiga University of Medical Science, FY2022) and by the NPO Super Scientist Program Plus.
Previous studies have reported a risk assessment model utilizing Himalayan region satellite imagery to estimate glacier lake outbursts from moraine-dammed lakes with high precision. This model considers factors such as topography, watershed area, estimated water storage based on glacial lake area, and downstream river gradient. However, it is not specifically designed for issuing alerts for dam failures, and therefore lacks the temporal resolution necessary to provide timely alerts before a collapse occurs. Recently, another research group demonstrated cases where glacial lake outbursts occurred within a few days of a significant increase in surface temperature (approximately 0.3°C/day) in the upstream glacier. The mechanisms of collapse involve various factors and regional dependencies, including glacial lake overflow due to meteorological factors, outbursts, avalanches of ice or rocks into the lake, among others. Consequently, a comprehensive understanding of crucial meteorological elements and the temporal resolution of dam failure estimation models is not yet sufficient to issue appropriate alerts.
This study aims to establish a high-precision glacial lake dam failure estimation method that can predict the collapse of moraine-dammed glacial lakes several days in advance. As the initial step towards this goal, we focused on the glacial lake area, which is believed to have a strong correlation with the collapse of moraine-dammed glacial lakes, and examined its correlation with the surface temperature of the upstream glacier. The study area includes the Shisper Glacier Lake in Hunza, Pakistan, and the Muchuwar Glacier, considered to be the water source for this glacier lake. Due to the larger area of the upstream glacier compared to the glacial lake, estimating the surface temperature of the glacier can be crucial for assessing the risk of collapse caused by the expansion of the glacial lake, even in situations where satellite image analysis is hindered by cloud cover over the glacial lake. This information is deemed essential for issuing timely alerts.
Analysis utilizes Landsat 7 and 8 satellite images (a total of 11 images) captured approximately one month prior to the reported Shisper Glacier Lake outbursts in 2019, 2020, and 2022. Surface temperatures of the Muchuwar Glacier were calculated from Landsat 7 band 6 (10.40-12.50 µm) and Landsat 8 band 10 (10.6-11.19 µm) images. The Shisper Glacier Lake area was determined from Landsat 8 visible composite images.
From the upstream edge of the glacier lake, a correlation analysis was conducted between the surface temperature of glaciers within a range of 2.5 km intervals, spanning 2.5 to 12.5 km. The results revealed a high correlation, with r values ranging from 0.65 to 0.90. These findings indicate that surface temperature is a precise indicator for predicting glacier lake outbursts, and the study discusses the potential application of these results to glacier lakes in other regions.
This research was partially supported by the Telecommunications Advancement Foundation under the "Development of Human Resource Education Methods with SDG Problem-Solving Capabilities Using Both ICT and Hands-on Approach (Shiga University of Medical Science, FY2022) and by the NPO Super Scientist Program Plus.