Fierr - Space Apps Challenge

Fierr| Spot That Fire V3.0

Awards & Nominations

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

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.

FIRE (Foreshadowing the Impact of fiREs)

Summary

Predict the economic damage of a potential fire by simply drawing the potentially burnt area on the map. Using US Census and previous wildfire data, the economic burden caused by a wildfire can be predicted.

How We Addressed This Challenge

In recent years, fires have been growing stronger, and occur more frequently than ever before. Unarguably, assessing the impacts of a fire after the fire has been completely contained is important, in order to improve and perform better the next time, and therefore minimize the economic impacts and more importantly the number of lives lost. However, there needs to be a way to predict at a certain point, what the impact of a hypothetical wildfire may be. This could be used to see which part of the fire needs to be contained first, and what the impact would be if the fire is not contained correctly.

This project provides the means to do this, through the prediction of costs of any potential fire. With the visual augmentation given by the active fire data on the map, the user can draw the shape of the fire as predicted by other services, and predict the economic impacts of the fire.

A mathematical model was created based on the population density, the mean house price, the area and the number of housing units, the area of the fire and other features, and fitted on past wildfires with reported financial impacts. Using this, a reasonably accurate model was created.

In order to help with prediction of wildfires, I have added an extra GIS layer, which includes the potential hazard of a wildfire occurring in the region. This is sourced from the US Forest Service. This, shows how dangerous a fire would be in each area and how likely a wildfire may occur. The inclusion of this in the map helps in the detection of possible wildfires in the future.

How We Developed This Project

Implementation details

Upon drawing a polygon, the data analysis starts, on the front end.
Random points within the polygon are sampled to find the proportion of the polygon that belongs to which county. In order to do this, the polygon is triangulated, the area of each triangle is found, and a cumulative distribution is created to make a uniform distribution of points on the polygon. After the coordinates of each point are found, they are queried into Nominatim Reverse Geocoding API, which returns their county.
For each relevant county, the total area of the county, the number of houses, the population density and the average house price are queried from their respective APIs
The data are fed into a model that predicts the total economic cost

Tech Stack

The solution has been implemented using JavaScript.
The Leaflet.js library has been used in order to display the map and the tiled FIRMS2 WMS service map.
The Leaflet.Draw library has been used to allow the drawing of polygons and shapes on the map.
There is no backend code written by me in this project. All of the data aggregation and calculations are done on the frontend

Future Improvements

Use Google Maps API for Reverse Geocoding, which returns the zip code of each coordinate. Therefore, smaller resolution can be had, which will improve the accuracy, since the assumption that the distribution of houses is uniform in each county will be reduced to the distribution being uniform in each ZIP code, which is a lot closer to being valid.
Find and train the model using more past fire data to improve accuracy
Automate the process of highlighting the region, so that the correct region is automatically selected
Show a separate report for the environmental damage caused by the fire (total area in acres of greenery burnt, forestry burnt, etc)

How We Used Space Agency Data in This Project

The current thermal anomalies are displayed on the map to help the user with drawing the correct area for the prediction of the fire from MOD14 and VIIRS.
Satellite imagery from past fires has been used to improve and evaluate the model and thus improve the accuracy of the model.