Hazard Modeling Using Sentinel Data for Risk Assessment and

Management of Tsho Rolpa Glacier, Nepal

Prasil Poudel, Nikesh Budha Chettri, Nabin Sah and Drabindra Pandit

Department of Physics, St. Xavier's College, Maitighar, Kathmandu, Nepal

Keywords: Glacier Lake Outburst Flood (GLOF), Synthetic Aperture Radar (SAR), Geographic Information System (GIS),

and Hazard Modeling.

Abstract: The accelerated melting of the Himalayan glaciers due to climate change has caused significant issues

including Glacier Lake Outburst Floods (GLOFs), to the downstream communities and the infrastructure.

This study focuses on the Tsho Rolpa Glacier in Nepal, utilizing Sentinel-1 Synthetic Aperture Radar (SAR)

data to measure ice velocity and analyze the impacts of glacier melting on lake dynamics. A GLOF occurs

when the volume and surface area of glacier lake water exceeds the capacity of the moraine dam. For hazard

modeling, SAGA GIS 8.3.0 was used for the terrain modeling and hydrological analysis, which highlighted

the flood-prone areas. ArcGIS 10.5 facilitated the integration of Sentinel-2 imagery with local topographic

data to predict the flood scenarios. The integration of these tools enhanced the accuracy of flood paths

predictions and provided much needed valuable insights into the impacts on the local infrastructure. The

results highlights the growing risks associated with the climate-induced GLOFs, demonstrating the

importance on real-time glacier monitoring and predictive hazard modeling. This study tries to be a helpful

tool for decision-support framework for mitigating the socio-economic impacts of GLOFs in vulnerable

regions such as the Himalayas.

1 INTRODUCTION

Glaciers are a very important component of the

hydrological cycle of the earth system i.e. glaciers

play a significant role as a fresh water reservoir in

maintaining global sea level and regional water

availability. Himalayan Glaciers acts as a freshwater

reservoir to rivers downstream, which acts as a

lifeline for millions of people living downstream.

However, in the past few decade due to the ongoing

impacts of the climate change, it has accelerated the

rate at which the glaciers are melting, increasing the

risk of Glacier Lake Outburst Flood (GLOF),

landslides and other environmental disasters that

threatens the local infrastructure, ecosystems and the

daily livelihoods of the people living in those GLOFs

prone areas.

In particular, the threat of GLOFs has raised a

significant concern to the local demographic, as these

events cause catastrophic flooding downstream when

the moraine dam holding back glacier lake water fails

due to excessive increase in the glacier lake due to

melting of the glacier lake. The accurate and timely

monitoring of glaciers and the hazards associated to it

would play a critical factor for the risk assessment and

disaster management in the event of GLOF incident.

Thanks to the advancement in satellite technology,

monitoring of glacier dynamics has become more

precise and accessible. The Sentinel-1 Synthetic

Aperture Radar (SAR) and Sentinel-2 optical data,

both provided by the European Space Agency (ESA),

have proven to be indispensable tools for glacier

monitoring and the development of hazard models.

While Sentinel-1’s radar capabilities enables the year-

round monitoring (even in challenging weather

conditions due to cloud penetrating radar capabilities)

and are ideal for measuring ice velocity which is a key

factor in understanding glacier dynamics (Copernicus

Climate Sentinel-1, n.d.) but Sentinel-2’s high-

resolution optical imagery is particularly useful for

assessing surface changes and developing detailed

hazard models and predict the flood-path and the

subsequent impacts of potential GLOFs.

Our approach to this study, was to generate

awareness towards the potential disaster than could be

caused by the GLOFs incidents in high impact areas

such as Tsho Rolpa Glacial Lake in the Rolwaling

Valley (Rounce, Watson, & McKinney, 2017). This

Poudel, P., Chettr i, N. B., Sah, N. and Pandit, D.

Hazard Modeling Using Sentinel Data for Risk Assessment and Management of Tsho Rolpa Glacier, Nepal.

DOI: 10.5220/0013288700003935

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 11th International Conference on Geographical Information Systems Theory, Applications and Management (GISTAM 2025), pages 165-171

ISBN: 978-989-758-741-2; ISSN: 2184-500X

165

study utilized the Sentinel-1 SAR data to measure the

ice velocity of the Tsho Rolpa Glacier from 2020 to

2023, enabling the analysis of glacier dynamics. In

addition to this, Sentinel-2 imagery was used to create

an extensive hazard model. This study serves as a small

yet significant step toward mitigating the impacts.

2 RELATED WORKS

2.1 Glacier Hazard Monitoring and

Risk Assessment

Glacier hazard modeling is very important for the

prediction of associated flood path in case of GLOF

incident. Various studies have already been published

on this topic, especially the effects of climate change

in glaciers (Bolch et al., 2012).

Studied glacier retreat and thinning pattern in

Himalayan glaciers. This study suggests the increase

in potential risk of GLOFs as ice volume declines.

2.2 Sentinel-1 SAR Applications in

Glacier Monitoring

Sentinel-1 SAR data provides all-weather year-round

glacier monitoring, which acts as a great tool to

monitor areas such as Himalayan region where cloud

coverage is common. Several glaciers in previous

studies such as (Rankl, Kienholz, & Braun, 2014)

have utilized Sentinel- 1 data to monitor changes in

Ice velocity and surface deformations.

2.3 Sentinel-2 Imagery Data for

Hazard Modeling

Since Sentinel-2 provides a higher spatial resolution of

10 meters for visible and Near Infrared band (NIR), 20

meters for red-edge and SWIR band and 60 meters for

atmospheric correction bound as compared to 5-20

meters of Sentinel-1 SAR (ChatGPT, Personal

Communication, 2024). Hence, Sentinel-2 can be an

important tool for monitoring small changes in glacier

flood path making it more effective in Hazard

modeling as compared to Sentinel-1 SAR (Paul et al.,

2016).

2.4 Glacier Lake Area and Volume

Change

The previous studies by (Maskey et al., 2013)

provides insights on the significant expansion of Tsho

Rolpa Glacier Lake, driven by impacts of climate

change and accelerated melting of its parent glacier.

The lake’s surface area, recorded at merely 0.23 km²

in 1957, increased substantially to 1.65 km² by 1997,

reflecting a dramatic growth of approximately 617%

over four decades. This expansion of Tsho Rolpa

Glacial Lake has heightened the threat of GLOFs.

3 METHODOLOGY

3.1 Study Area

Tsho Rolpa glacier, located in Rolwaling Valley of

Himalayan region of Nepal is prone to GLOF

incidents because of its large glacier lake that is

actually growing because of melting glaciers

(Damen, 1992) This study analyzes the spatial and

temporal aspects of glaciers and glacier lakes.

3.2 Data Acquisition

Our study utilizes Sentinel-1 and Sentinel-2 data

obtained from the open-access Copernicus Open

Figure 1: Google. (2021). [Map showing Tsho Rolpa Glacier] [Map]. Google Earth Pro (Version 7.3). Imagery date: March

25, 2021. Retrieved January 20, 2025, from https://earth.google.com/.

GISTAM 2025 - 11th International Conference on Geographical Information Systems Theory, Applications and Management

166

Access hub (Copernicus.eu). Since Sentinel-1 data

were used for different purpose like monitoring

natural disasters, environmental monitoring and

different climatic analysis

(Copernicus Sentinel-1, n.d.). In our study, we

have use Sentinel-1 as well as Sentinel-2 data for

analyzing Ice velocity and Hazard modeling

respectively provided by the European Space Agency

(ESA). The datasets were taken in following ways:-

3.2.1 Sentinel-1 Data

Sentinel-1 uses a spatial resolution of 5m in range

with 20m in azimuth (Du et al., 2021). The Twin

polar-orbiting satellites i.e Sentinel-1A and Sentinel-

1B are equipped with supply geographical data for

environmental and security warranting for the

expansion of the remote sensing, world economy and

business which means the satellites function both

during the day and at night and perform a synthetic

aperture with radar imaging i.e. it process radar signal

to create high-resolution image of Earth which is

obtained by simulating a large synthetic aperture

using the motion of the radar platform. We can obtain

the imagery using Sentinel-1 bands in any weather.

An active phased array antenna called C-SAR was

developed to offer faster azimuth and elevation

scanning which allows to cover bigger areas of

incidence angle to support the SAR operation

(Yulianto et al., 2021). In this research, High-

resolution SAR images from the Sentinel-1 satellites

were downloaded in Ground Range Detected High

resolution (GRDH; Level-1).

Sentinel-1 is a sun- synchronous, near-polar orbit

satellite with 175 orbits around the earth and 12 days

repeat cycle. Both Sentinel-1A and Sentinal-1B share

the same orbit plane with a 180- orbital phasing

difference. The orbit altitude of Sentinel-1 Satellite is

693 km (Yulianto et al., 2021, September).

We have acquired the imageries in Interferometric

Wide Swath (IW) Mode, which helps to monitor large

area, and in Vertical-Horizontal (VH) Polarization i.e

the signal is transmitted vertically and received

horizontally. Also, all the images were downloaded in

GRD.

3.2.2 Sentinel-2 Optical Imagery Data

Sentinel-2 provides a multispectral imaging that

allows for high-resolution imagery data, useful for

analyzing surface characteristics, glacier extent etc.

For this study, Sentinel-2 imagery data was assessed

from Copernicus Open Access hub/Copernicus

browser, which is a platform for Sentinel data. A

processing level of 2A was used, which provides

atmospherically correct surface reflectance

(Copernicus Open Access Hub, n.d.)

3.3 Methods of Hazard Modeling

Hazard Modeling is crucial for predicting the

potential flood paths in case of GLOF incidents.

Knowing the potential flood path/flow paths would

prepare local governmental bodies and demography

to be more prepared in such incidents. Along with that

it is also crucial for the planning infrastructural

development around such GLOF prone areas.

For hazard modeling Sentinel-2 imagery data was

utilized which was taken from the Copernicus hub

and then pre-processed using SAGA GIS 8.3.0

software’s “Filled sink (Wang & Liu)” pre-

processing tool to correct the elevation layer in

Digital Elevation Model (DEM) map and from DEM

map, channel network vector was created using the

Hydrology tool in Terrain analysis of SAGA GIS

8.3.0. Then the original DEM map and channel

network was imported into ARCGIS 10.5 and merged

into a single layer. The channel of order greater than

0.1 in terms of width and length was specified and

connected creating the hazard modeling layer in 2D.

Now the map was converted into a KML file and 3D

visualization was done in Google earth. Both the

DEM model and TIN visualization layers in Google

Earth Pro are taken as final results as 2D and 3D

hazard models. Both DEM layer and TIN layer were

used as results as they both could high light the aspect

of elevation to present the actual terrain features.

3.4 Ice Velocity

Glacier ice velocity plays a crucial role in understanding

Table 1: Parameter of images.

Acquisition date Acquisition date Acquisition date Acquisition date

2020-01-04 2021-01-10 2022-01-05 2023-01-12

2020-02-09 2021-02-15 2022-02-10 2023-02-17

Hazard Modeling Using Sentinel Data for Risk Assessment and Management of Tsho Rolpa Glacier, Nepal

167

Figure 2: 2D DEM layer Hazard Model.

Figure 3: 3D Hazard Model.

the overall glacier dynamics influenced due to climate

change. The velocity of glaciers caused by different

factors including geographical location, ice thickness,

and environmental conditions (Hyde, 2024). In regions

like Himalayans the glacier dynamics affects the

availability of fresh drinking water highlighting socio-

economic effect to the country and measurement of

glacier’s ice velocity are essential for modeling future

glacier changes and potential natural calamities like

flood and landslide (Millan et al., 2023). Ice Velocity of

glaciers can be measured using different techniques but

in this study we have used optical and SAR imageries to

analyze the glacier ice velocity since high resolution

optical imageries provides high accurate measurement

(Gu et al., 2024). In this research study, we have use

offset tracking analysis calculated by Sentinel

Application Platform (SNAP) to analyze image intensity

information for identifying and matching points

between image (Gu et al., 2024). In offset tracking

random points are generated and compared between two

points which allows to analyze glacier movement with

high accuracy and calculated by

Offset =

√



𝐴



𝐷





(1)

Where, Offset is the offset of the drifted glacier

area, A is the average value of offset drifted glacier

area, D is the standard deviation of the drifted glacier

area (Gu et al., 2024).

We have used the following technique for the

analysis of the Ice velocity:

GISTAM 2025 - 11th International Conference on Geographical Information Systems Theory, Applications and Management

168

Figure 5: Flowchart of Ice Velocity Processing.

4 RESULTS AND DISCUSSION

4.1 Results from Hazard Modeling

The Hazard Modeling of Tsho Rolpa glacier lake

reveals detailed potential flood paths and hazard zones

in case of a GLOFs event. This model shows flood

paths (purple and blue lines as in figure 2 and figure 3)

where water could flow as a direct result of an outburst

event of the Tsho Rolpa glacier lake. From this model,

it was clearly shown that rural communities such as Na

Village (as seen in map) are at a high risk which serves

as a critical tool for identifying vulnerable areas and

guiding mitigation strategies.

4.2 Results from Ice Velocity

Measurement

From the figure 5(a), (b), (c), (d) it was found that the

ice velocity in the area in the year 2020 January-

February was a maximum of 0.129m/day and

minimum of 0.016m/day indicated by Red color and

White color respectively similarly in the year 2021

January-February there was a slightly fluctuation in

the velocity with a maximum velocity of 0.056m/day

and minimum of 0.002m/day indicated by Red color

and White color respectively. In the year 2022, we

can observe that the ice velocity became twice as of

2021 with a maximum velocity of 0.106m/day and

minimum of 0.006m/day indicated by the red and

white color. In the year 2023, we observed a decrease

in the velocity of ice with maximum of 0.058m/day

and minimum of 0.002m/day.

5 DISCUSSION

5.1 Interpretation of Hazard Modeling

Results

The Hazard Modeling of Tsho Rolpa glacial lake

highlights the GOLF risks of the region. The flood

paths visualized in the Figure 2 and Figure 3 shows

those rural communities such as Na Village which are

in a direct hazard zone. The visualized flood paths

demonstrate the utility of integrating remote sensing

data with the hazard assessment tools. The accuracy

of the flood paths can be further refined which was

discussed in the future works.

5.2 Interpretation of Ice Velocity of

Measurement

From the above result, we can conclude that the Ice

velocity in the area has fluctuation over time and can

lead to moraine dam collapse due to continuous flow

of water in the Tsho Rolpa lake and with fluctuation in

ice velocity the possibility of GLOF can be occurred

anytime. With Global temperature rising (Copernicus

Climate Change Service, n.d.), the possibility of

glacier melting with numerous amount of water

flowing

into

the

lake

can

lead

possible

GLOF

Hazard Modeling Using Sentinel Data for Risk Assessment and Management of Tsho Rolpa Glacier, Nepal

169

Figure 5(a): Ice velocity of 2020.

Figure 5(b): Ice velocity of 2021.

Figure 5(c): Ice velocity of 2022.

Figure 5(d): Ice velocity if 2023.

condition in the area which can badly affect the

country’s economic as well as long-term effect can be

seen.

6 CONCLUSION AND FUTURE

WORKS

GLOFs are natural disasters that will happen

eventually. The only viable method of risk reduction

are the early warning systems. The study of glacier

ice velocity provides us an insight about the dynamics

of glaciers, sea-level rise and response to climate

change. With the help of ice velocity integrated with

hydrology, we can design an early warning system.

The hazard modeling provides crucial insights on the

nature of flood paths and high-risk zones, where the

early warning can effectively reduce the risk on life

and properties. The rapid expansion of glacial lake

and the resulting potential flood path emphasizes the

urgent need for intervention for the mitigation of

GLOFs.

One of the major challenges many countries

that has direct access to the ocean or sea are facing is

continuous sea-level rise and one of the major cause

for sea level rise is melting of glaciers, which add tons

and tons of water continuously in the ocean resulting

in rise of sea level. Future work should focus on

refining the accuracy of flood paths generated by

hazard models by incorporating real-time data.

Further works can be done analyzing different

parameters like ice thickness, surface displacement

using satellite data as well high precision GPS

stations and InSAR to get high accuracy results.

Similarly more sophisticated modes can be developed

that can simulate ice velocity in different weather and

climate variables taking in note about the surface as

well as internal glacier melting and deformation. The

current hazard model relies on the geospatial data and

the assumption of the lake stability which might not

full capture the dynamic nature of GLOF events so

direct field-based data collections and validation of

observations along with advanced hydrodynamic

modeling could address these gaps and provide a

more refined hazard model.

GISTAM 2025 - 11th International Conference on Geographical Information Systems Theory, Applications and Management

170

REFERENCES

Bolch, T. K. (2012). The state and fate of Himalayan

glaciers. Science, 336(6079), 310–314.

doi:https://doi.org/10.1126/science.1215828

ChatGPT. (2024). Explanation on spatial resolution of

Sentinel-2 and Sentinel-1.

Copernicus Climate Change Service. (n.d.). Home Page (or

Copernicus Climate Change Service Website).

Retrieved from European Centre for Medium-Range

Weather Forecasts: https://climate.copernicus.eu/

Damen, M. C. (1992). Study on the potential outburst

flooding of Tsho Rolpa glacier lake, Rolwaling valley,

East Nepal. International Institute for Geo-Information

Science and Earth Observation.

Du, Q. L. (2021). Deformation Monitoring in an Alpine

Mining Area in the Tianshan Mountains Based on

SBAS-InSAR Technology. Advances in Materials

Science and Engineering, 2021(1), 9988017.

doi:https://doi.org/10.1155/2021/9988017

Gu, F. Z. (2024). The velocity extraction and feature

analysis of glacier surface motion in the Gongar region

based on multi-source remote sensing data. Frontiers in

Earth Science, 12(Article 1413531).

doi:https://doi.org/10.3389/feart.2024.1413531

Hyde, A. (2024, January 15). Measuring glacier velocity.

Retrieved from Antarctic Glaciers:

https://www.antarcticglaciers.org/glaciers-and-

climate/observing-and-monitoring-glaciers-and-ice-

sheets/measuring-glacier-velocity

Maskey, R. K. (2013). Water and energy security in the

mountainous region of Nepal: Lessons from Dig-Tsho,

Tsho-Rolpa, and Imja glacial lakes.

Millan, R. M. (2023). Author correction: Ice velocity and

thickness of the world’s glaciers. Nature Geoscience,

16(1), 188.

Paul, F., Winsvold, S. H., Kääb, A., Nagler, T., &

Schwaizer, G. (2016). Glacier remote sensing using

Sentinel-2. Part II: Mapping glacier extents and surface

facies, and comparison to Landsat 8. Remote Sensing,

8(7), 575. doi:10.3390/rs8070575

Rankl, M. K. (2014). Glacier changes in the Karakoram

region mapped by multimission satellite imagery. The

Cryosphere, 8, 977–989. doi:https://doi.org/10.5194/tc-

8-977-2014

Rounce, D., Watson, C., & McKinney, D. C. (2017).

Identification of hazard and risk for glacial lakes in the

Nepal Himalaya using satellite imagery from 2000–

2015. Remote Sensing, 9(7), 654.

doi:https://doi.org/10.3390/rs9070654

Yulianto, S. A. (2021, September). Spatial Distribution of

Paddy Growth Stage Using Sentinel-1 based on CART

Model. 2021 IEEE Asia-Pacific Conference on

Geoscience, Electronics and Remote Sensing

Technology (AGERS). Institute of Electrical and

Electronics Engineers.

doi:https://doi.org/10.1109/AGERS53903.2021.96173

Google, E. (2021). Google Earth Pro [Map].

Hazard Modeling Using Sentinel Data for Risk Assessment and Management of Tsho Rolpa Glacier, Nepal

171