Lagunas| A Flood of Ideas
A Flood of Ideas
Flood Prediction by previous floods
Summary
Our algorithm will analyze the previous floods' data about the probability of that flood happening and the damages that it could cause, then it will compare this data with the one we have about the current meteorology, so that we will estimate the damages and the costs that it will cause.After that we will store the data and after the flood happens we will compare to our predictions for improving our algorithm. We will be using KMeans algorithm.
How We Addressed This Challenge
We created an algorithm that will predict floods with enough time for people to get to a safe place, we will also estimate the damage and the costs that the flood might cause.
Our algorithm will check all the current data and compare it with previous floods, then it will calculate the probability of a flood happening
Our motivation comes from wanting to inform about the danger of floods because they create a large amount of economic and personal damage, for example if a flood spreads to some crops, they spoiled both the harvest and said fields, which would create a problem both food, as economic for the farmers who were cultivating those lands, or for example if the flood occurs in a city, it would affect all the people living in that city, with the flooding of their garages, and it would be a problem both economic and personal, since people could not go to their jobs, which would not generate an economy in that city for that time, and they would lose many material goods. That is why we have become aware, and we have arranged to carry out this project
How We Developed This Project
- Probability of the flood happening
- Damage it would cause
- Goals
We will consider the following data: Season, Amount of rainfall, duration, intensity and area of the storm, ground temperature, topography and vegetation. We got this data from JAXA GLOBAL RAINFALL WATCH (GSMaP), Global Flood Monitoring System (GFMS), Canadian Space Agency Open Data Portal.
With Kmeans algorithm we will compare this data of the current case with the dataset getting a probability of a flood.
After predicting if a flood is going to happen we are going to predict the damage considering the following data: with the data of the previous point we’ll also add dead people, displaced people and get an estimation of it.
For coding this we are going to use Matlab and bring a small example of how it would work, for simplicity we are going to create our own small dataset with data from the space agency resources.
Our dataset has the following data:
Season; Amount of rainfall; duration; intensity; gorund temperature; topography; vegetation.
Our goals here are to predict floods with a good chance of guessing and to prevent people of getting harmed or even die.
How We Used Space Agency Data in This Project
We used JAXA GLOBAL RAINFALL WATCH (GSMaP) and Canadian Space Agency Open Data Portal to get data of the floods like amount of raing, duration, intensity and area of the storm.
We used Global Flood Monitoring System (GFMS) and Today's Earth for the topography also we used "Technical Guidance for Monitoring and Reporting on Progress in Achieving the Global Targets of the Sendai Framework for Disaster Risk Reduction" to bear in mind the best data to consider.
Project Demo
Project Code
Data & Resources
JAXA GLOBAL RAINFALL WATCH (GSMaP)
Canadian Space Agency Open Data Portal
Global Flood Monitoring System (GFMS)
Today's Earth
"Technical Guidance for Monitoring and Reporting on Progress in Achieving the Global Targets of the Sendai Framework for Disaster Risk Reduction"