naderbekir88/nasa-space-competition

Landslides occur in all 50 states. Causing between 25 and 50 deaths every year in the US alone and thousands more in vulnerable areas around the globe. Landslides include debris avalanches, debris flows down carrying large rocks, block movements causing huge damage and loss to people, their homes, farms and cattle, industrial and other structures. Our project detects landslides' whereabouts and predicts their level of risk, potential casualties, and the cost of their impacts. Using a user-friendly website &machine learning model trained on data from NASA satellites and ground-based sources. providing affected communities and governments with the information needed for timely decision-making.

Identifying the risk of landslides challenge

Overview

Landslides, a global hazard causing numerous deaths and extensive damage, affect all US states and vulnerable regions worldwide. Our project aims to detect and assess landslides using a machine learning model trained on NASA satellite and ground-based data. The model predicts the risk level, potential casualties, and impact costs of landslides. Results are presented through a user-friendly website, empowering affected communities and governments with timely information for effective decision-making.

Project Demo

Believers.Project.Demo.mp4

How We Addressed This Challenge

Our team has developed an innovative machine learning model to predict landslides and assess their potential impact. The model was trained on a comprehensive dataset sourced from NASA's satellites, including wind, rainfall, and earthquake data.

We utilized the Integrated Multi-satellite Retrievals (IMERG) algorithm, which combines information from two key sources:

1. Early precipitation estimates were collected by the Tropical Rainfall Measuring Mission (TRMM) satellite between 2000 and 2015.
2. Recent precipitation estimates gathered by the Global Precipitation Measurement (GPM) satellite from 2014 to the present.

By leveraging this rich dataset, our model can predict the likelihood and severity of landslides in various regions.

The culmination of our project is an interactive, user-friendly website that serves as a valuable resource for both researchers and the general public. The website features our training datasets, clear and concise analysis, conclusions, and visualizations.

For researchers and scientists, this platform facilitates the integration of our findings into their processes, enabling them to build upon our work and advance the field of landslide prediction and mitigation.

Simultaneously, the website provides individuals an accessible tool to determine the risk of experiencing a landslide. By inputting their location and other relevant factors, users can quickly and easily assess the potential danger they face and take appropriate precautions.

Significance of Our Landslide Forecasting Project

• Equipping governments with timely forecasts facilitates proactive decision-making and risk mitigation strategies.
• Empowering rural communities with knowledge about their potential landslide risk, enabling them to take appropriate precautions.
• Providing governments and communities with sufficient lead time to prepare for and respond to potential landslide events effectively.
• Contributing to preserving the natural environment by minimizing the devastating impacts of landslides through early warning systems.
• Supplying authorities with robust scientific and statistical data to support evidence-based decision-making processes related to landslide risk management.
• A base for researchers to build upon their studies and investigations.

How the tool works

It allows the user to easily specify a location and time in the website heat map, the map then shows if there have been previous landslides in that place, and then we provide the user with information about the previous damage caused in that particular structure, the extent of it, and finally the model makes a forecast of the upcoming landslide date and.

Tools / Coding Languages/ Software Used:

• Python
• HTML
• CSS
• Pandas
• Numpy
• Scikit-learn

Project Link

Believers - Identifying The Risk Of Landslides