Spotifyre

Spotifyre| Spot That Fire V3.0

Awards & Nominations

Spotifyre has received the following awards and nominations. Way to go!

Local Peoples' Choice Winner

Global Nominee

Spot That Fire V3.0

Recent wildfires worldwide have demonstrated the importance of rapid wildfire detection, mitigation, and community impact assessment analysis. Your challenge is to develop and/or augment an existing application to detect, predict, and assess the economic impacts from actual or potential wildfires by leveraging high-frequency data from a new generation of geostationary satellites, data from polar-orbiting environmental satellites, and other open-source datasets.

Summary

Forest fires are becoming more severe and damaging as the effects of global climate change accelerates. In Canada, forest fires have burned millions of hectares and cost us billions of dollars in damages over the past decade.Our contribution to fighting these fires is Spotifyre. This web application uses public data from MOPITT, Copernicus, and various NASA satellites to monitor atmospheric conditions and report high fire risks, or potential fires throughout Canada. It also allows the user to submit their own sightings of fire hazards, so that everyone can help fight forest fires.

How We Addressed This Challenge

we developed a web application that will visualize data to predict potential fires across Canada. Using CSA and Nasa data, the solution includes a predictive analytic model that can assess regional risk levels of wildfire occurrence. The web application implements a Crowdsourcing model that allows users to report sighting/incidence of fire activity in real time.

The web application is important because it will help us to predict wildfire trends and patterns. This will result in being able to determine faster course of action to help with preventing risk associated to wildfires.

How We Developed This Project

A big reason why we decided to choose this project was because of the recent wildfires that occurred around the world. We believe that this issue is pressing. Also predictive analytics is a relatively new field that our group wanted to explore. With access to the Nasa and CSA datasets, we believed that we could produce useful intelligence to help predict wildfire patterns.

We approached this project by first prioritizing what datasets we had access to and understand what could potentially cause a wild fire. To best utilize the data, we had to research what data correlated and aligned with our goals. After, we filtered through the data and split up the team into front-end and back-end developers. The front end would be in charge of the data visualization and setting up the servers. While the back end would filter through the datasets, and then feed the data into a predictive analytic model to train it. The model would then be used to analyze future data and predict trends in wildfire.

We used React.js for the front end and python for back-end.

Problems: We did not realize how much data needed to be cleaned and transformed into the proper format so that all streams datasets can be merged without conflict.

For the front-end, we were using amchart4 api to do the data visualization and it was our first time using that api. We had to spend a lot of time reading and understanding the documentation so it slowed down the progress.

Another issue was that we did not enough time to build the model, we were only able to modify all the data and process it so it's ready to be used to train a model.

Achievements: We managed to get a front-end running with some mock data, and we were able to modify all the data and process it. Also we were able to get some sleep in between so I will also count that as a win.

How We Used Space Agency Data in This Project

All of our data used in this project comes from public data provided by MOPITT, NASA, and the ESA's Copernicus program. These data sources provide detailed atmospheric conditions, such as carbon monoxide concentrations that are key indicators of forest fire activity.

We used this data to train a model that would identify any potential fires throughout Canada, as well as determining what regions are at an elevated risk for forest fires.