SMART INDIA HACKATHON 2024
Problem Statement ID : SIH1650
Problem Statement Title : Early Warning System for Glacial
Lake Outburst Floods (GLOFs)
Theme : Disaster Management
PS Category : Software
Team ID : SIH1023
Team Name : Team Slytherin
TEAM
SLYTHERIN Early Warning System for Glacial Lake Outburst
Floods (GLOFs)
Glacial Lake Outburst Floods (GLOFs) pose a severe threat to communities in mountainous regions due to climate change
accelerating glacial melt. Current monitoring methods are limited, leading to delayed warnings and increased risks to lives,
infrastructure, and the environment.
PROPOSED SOLUTION :
We propose an advanced Early Warning System (EWS) that integrates satellite imagery and IoT sensors (seismometers, radar level
sensors, temperature sensors, ultrasonic water flow meters) to continuously monitor glacial lakes. By using machine learning
algorithms, the system will predict potential GLOF events based on changes in water levels, ground movement, and environmental
conditions, providing timely warnings for evacuation and mitigation.
INNOVATION & UNIQUENESS:
Our solution combines satellite data with on-site IoT sensors to trigger real-time monitoring only when significant environmental changes are
detected. This reduces unnecessary data collection and improves prediction accuracy. The system’s ability to provide early, precise alerts,
coupled with its scalable design for different glacial environments, makes it an innovative and cost-effective tool for disaster risk reduction.
ML
Feed Predict
outburst
Data ML model Alert and Rescue GLOF
TEAM
SLYTHERIN
TECHNICAL APPROCH
TECHNOLOGY STACK
Seismometer Radar level Temperature sensor Apache Spark Google cloud
Ultrasonic water flow SAR & Google Earth engine TensorFlow SNS
Monitor any seismic
sensor
meter
activity that could Measure water Measure water
temperature at To measure flow
indicate ground levels without Processing satellite For large-scale data To train and test Alerting through Used for storing
various depths rates and detect
movement and Dam direct contact imagery. processing. the ML model SMS, emails, or the data.
sudden increases
stability telphonic.
Inputs for prediction
Preprocess data
Pile up data Alert alarms
Missing data
Noisy data for warning
Inconsistent data
Snow depth Lake size
Incorrect imputation
Data Transmission
Class Imbalance Check for anamoly
Frequent monitoring
Notifying
Aggregate data
Risk analysis and Hazard identified
decision making people through
----------------------------- SMS
Zoning
WORK FLOW Train model by feeding
Lake outburst imagery Periodic landscape capture
accumulated data
Data feed from satellite
Machine learning model
Accumulate data Signal the
rescue team
Create
homogenized
Aggregate data database for model
Temperature Seismic graph Water level
Snow depth training Train model
Fluctuation captured by sensors Alert and Rescue
TEAM
SLYTHERIN
FEASIBILITY AND VIABILITY
FEASIBILITY POTENTIAL CHALLENGES AND STRATEGIES FOR OVERCOMING
RISKS CHALLENGES
The project leverages
proven technologies like Sensors may Use high-quality, weather-
satellite imagery, IoT malfunction due to resistant sensors, establish
sensors, and established environmental factors routine calibration
machine learning models.
and disrupt data
These components are
transmission Begin with a pilot project in a
widely used in various
And high initial costs high-risk area to prove the
industries, ensuring that
the technical
pose funding challenges. system's value and attract
implementation is further funding.
achievable.
TEAM
SLYTHERIN IMPACT AND BENEFITS
[Link]-Saving Alerts: Provides early warnings for GLOFs, [Link] Public Trust and Engagement: Building a reliable
giving communities critical time to evacuate and reducing the EWS increases public trust in local governments and scientific
potential for loss of life. agencies, encouraging greater public engagement and adherence
to safety protocols, which is crucial for the effective
implementation of emergency measures.
[Link] to Scientific Research: The data collected from
the sensors and satellite imagery can be used for further
scientific research on glacial dynamics and climate change,
contributing to broader knowledge and future advancements in
disaster prediction and management.
[Link] Planning : The system provides actionable insights
and real-time data to local authorities, enabling them to make
faster and more informed decisions about evacuation routes,
resource allocation, and emergency responses, thus improving
the overall efficiency of disaster management operations.
TEAM
SLYTHERIN
REFERENCES AND RESEARCH
[Link]
_potential_glacier_lakes_A_case_study_in_Sikkim_Himalaya
[Link]
[Link]
systems/?gad_source=1
[Link]
[Link]