Conferences
Conference Papers
European Geophysical Union (EGU) General Assembly 2019 at Vienna, Austria from April 07 - 12, 2019.
Title: Assessing Glacial Lake Outburst Flood Risk in the Upper Teesta River Basin of Eastern Himalaya, India.
References:
-
Quincey, D.J., Richardson, S.D., Luckman, A., Lucas, R.M., Reynolds, J.M., Hambrey, M.J., and Glasser, N.F. 2007. Early recognition of glacial lake hazards in the Himalaya using remote sensing datasets. Global and Planetary Change. 56, 137-152.
-
Watson, C.S., Carrivick, J. and Quincey, D. 2015. An improved method to present DEM uncertainty in glacial lake outburst flood propagation using stochastic simulations. J. Hydrol. 529, 1373-1389.
4th World Congress on Disaster Management (WCDM) at the Indian Institute of Technology (IIT), Mumbai from Jan 29- Feb 01, 2019. [details]
Title: Modelling the outburst flood from Lhonak Glacial Lake of Sikkim Himalaya for GLOF risk assessment using GIS and HEC-RAS.
Abstract: ‘Himalaya’ is known as the third pole and the water tower of Asia which is the home of many natural hazards. Glacial Lake Outburst Floods (GLOFs) are among the most gigantic natural hazards posing significant threats to the livelihoods and infrastructures downstream. This study assesses the potential downstream risk from the Lhonak glacial lake by modelling the GLOF using GIS based Monte Carlo Least Cost path (MC-LCP) method and compare the model output with more stochastic two dimensional (2D) HEC-RAS model. The Lhonak Lake (27°54'45.14"N, 88°11'32.97"E) is located at an elevation of 5218m and is expanding drastically over few decades. High resolution multispectral satellite images of spatial resolution 5.8m (Resourcesat-1, LISS-IV) for two different time periods (2018 and 2011) and the ALOS PALSAR Digital Elevation Model (DEM) data of spatial resolution 12.5m, were processed in Erdas Imagine and ArcGIS software environments to derive the required information for modelling. The MC-LCP model was then executed at three different iterations (50, 100 and 500) for simulating the potential GLOF scenarios. The latest glacial lake inventory database of January 2018 revealed that it covered an area of 1.3 km2 which had increased by 0.2 km2 from 2011, probably due to an overall temperature rise across the entire Teesta River Basin, inducing further glacial melt, and held an estimated 51054972 m³ volume of water. The MC-LCP model output showed that an area of 11.9 km2 was inundated at 50 iterations, which increased to 14 km2 and 15.7 km2 at 100 and 500 iterations respectively. The model run with 100 iterations was considered for assessing GLOF risk and compared with the HEC-RAS model output. At 100 iterations, the lake had the potential to generate a 73.4 km long GLOF event downstream, which would inundate an area of 14 km2 and adversely affect 43 buildings (covering 0.02 km²), 3.2 km length of roads and 1.4 km² and 9 km² of vegetation cover and open land respectively. Whereas, the HEC-RAS flood inundation model showed that a similar GLOF event (73.7 km long) would inundate 23.8 km² of total catchment area and affect 113 buildings (0.01 km²), 14.9 km road lengths, 5.4 km² of vegetation cover and 13.8 km² of open land. Based on the cumulative score of all possible risks of a GLOF event in an outburst scenario using the GIS based MC-LCP model, this lake was considered to be highly hazardous, requiring urgent action to mitigate its high potential risk. It was also noticed that the newly developed MC-LCP method enabled an improved flow routing scenario and was capable of generating acceptable flood extents, even in a region of complex topography, enhancing its status as a significant first-pass hazard assessment tool for flood inundation mapping in the data poor Himalayan region. Such output maps can then be disseminated among local disaster management bodies for risk awareness and mitigation. In conclusion, further research is required to develop an early warning system using more stochastic models to mitigate the risks related to any GLOF events from this lake.
7th International Conference on Water and Flood Management (ICWFM) at Bangladesh University of Engineering and Technology (BUET), Dhaka from Mar 02-04, 2019. [details]
Title: A GIS-based Monte Carlo Least Cost Path (MC-LCP) method for assessing the outburst flood risk from Phunri Glacial Lake, Sikkim Himalayas, India.
INTRODUCTION:
Glacial Lake Outburst Floods (GLOFs) are one of the most potent natural hazards in the Himalayan region, posing significant threat to the millions residing downstream from their occurrence sites along the region’s foothills and extensively threatening built infrastructure and livelihoods. This paper examines the efficacy of a GIS-based Monte Carlo Least Cost Path (MC-LCP) method for assessing the GLOF induced risks from the Phunri Lake, which is located in the source region of the Teesta River, an important right bank tributary of the River Brahmaputra. The Lake is located at 27°59'26.62"N and 88°48'55.90"E, in the eastern Indian state of Sikkim and previous studies have identified it as a potential GLOF site.
METHODOLOGY
High resolution multispectral satellite images of spatial resolution 5.8m (Resourcesat-1, LISS-IV) for two different time periods (2018 and 2011) and the ALOS PALSAR Digital Elevation Model (DEM) data of spatial resolution 12.5m, were processed in Erdas Imagine and ArcGIS software environments to derive the required information for modelling. The GIS based MC-LCP model developed by Watson et al., (2015) was then executed at three different iterations (50, 100 and 500) for simulating the potential GLOF scenarios.
RESULT AND DISCUSSION
The latest lake inventory database (of January 2018) revealed that it covered an area of 1.8 km2, which had increased by 0.1 km2 from 2011, probably due to an overall temperature rise across the entire Teesta River Basin (as substantiated by climatic datasets for the area), inducing further glacial melt, and held an estimated 76711132.9 m³ volume of water. The model runs showed that an area of 18.8 km2 was inundated at 50 iterations, which increased to 22.8 km2 and 29.3 km2 at 100 and 500 iterations, respectively. At 100 iterations, the Lake had the potential to generate an 89.1 km long GLOF event downstream, which would inundate an area of 22.8 km2 and adversely affect 205 buildings (covering 0.07 km2), 31.8 km length of roads and 1.8 km2 of vegetation cover, despite it being located 91.7 km upstream of the basin outlet.
Based on the cumulative score of all possible risks of a GLOF event in an outburst scenario, this lake was considered to be highly hazardous, requiring urgent action to mitigate its high potential risk. It was also discerned that the newly developed MC-LCP method enabled an improved flow routing scenario and was capable of generating acceptable flood extents, even in a region of complex topography, enhancing its status as a significant first-pass hazard assessment tool for flood inundation mapping in the data poor Himalayan region. The study also highlighted some drawbacks of this model, in that it lacked a physical basis for the performed simulations and showed that at 500 iterations, the flood inundation probability (confidence level) was only 25%, as the model could successfully run only up to 127 times, possibly due to its inability to process a high resolution DEM dataset.
CONCLUSION
Further research is thus required to develop a more stochastic model for generating more robustly modelled scenarios. Such output maps can then be disseminated among local disaster management bodies for risk awareness, mitigation and forecasting of GLOF hazard which can potentially save many lives and properties.
REFERENCES
Watson, C.S., Carrivick, J. and Quincey, D. 2015. An improved method to present DEM uncertainty in glacial lake outburst flood propagation using stochastic simulations. J. Hydrol. 529, pp. 1373-1389.
