High-Level Project Summary
Team EQsight noticed valuable Earthquake Catalog API data openly provided by USGS and decided to upgrade existing Earth data visualization app - “QuakeHunter”. Given GeoJSON data could be used as point-of-reference for advanced predictive models, anomaly detection and distributional analysis. After reviewinig QuakeHunter repository team focused on refactoring Earthquakes Activity time-series graphic and updated algorithm to correctly count daily earthquakes for given coordinates point and radius. Additional feature was added to find maximum daily magnitude in Richter scale to test the hypothesis if increased number of earthquakes could distinguish higher magnitude earthquake.
Link to Final Project
Link to Project "Demo"
Detailed Project Description
Objective: Improve an existing Earth data visualization application by adding data analysis features to it.
- Team EQsight noticed valuable Earthquake Catalog API data openly provided by USGS and decided to upgrade existing Earth data visualization app - “QuakeHunter”.
- After reviewing QuakeHunter repository team focused on refactoring and implementing additional features for the Earthquakes Activity graphic.
- With added changes it could be possible to test the hypothesis: If increased number of daily earthquakes for given coordinates point in 1000 km. radius could distinguish higher magnitude earthquake?
Update Summary
- Radius Search functionality was moved to separate tab to implement geolocation feature to automatically input your device coordinates for Earthquakes Activity analysis.
- Refactored algorithm to correctly count number of daily earthquakes for given coordinates point and radius.
- Added feature to find maximum daily magnitude in Richter scale.
- Updated Earthquakes Activity graphic UI to improve data analysis.
- Proving hypothesis needs additional layers of data like earthquake depth and distance from recording station.
Space Agency Data
For this challenge we used data from Earthquake Catalog API from United States Geological Survey.
Hackathon Journey
It was really interesting to better understand science behind earthquakes observation and prediction. Also, it was surprising to easily access and explore data in Earthquake Catalog API, very important dataset to build predictive models. Space Apps hackathon organized participants to connect on Discord before event starts to enable early team formation. EQsight team seemed motivated but during event couldn't ensure communication during development time on weekend. Initial plan was to create proof of concept predictive model, but limited time forced to pivot idea couple times. The final decision was to individually focus on refactoring JavaScript code to update Earthquakes Activity time-series graphic.
References
https://github.com/WorldWindLabs/Quake-Hunter
https://github.com/WorldWindLabs/EarthquakeApp
https://earthquake.usgs.gov/fdsnws/event/1/
https://plotly.com/javascript/
Tags
#USGS, #earthquake, #earth-data-visualization, #Richter