SpaceBox

Roorkee | Save the Earth from Another Carrington Event!

Forecasting the upcoming Carrington event with thw help of previous data from reliable souces.

High-Level Project Summary

The challenge was to predict the onset of next carrington event based on data from the earlier occurence. The data from NASA was abrupt and a bit off-key to predict, hence we did an extensive pre-processing on data available. The basic idea of our team in this project was to set the data right according to our understanding, make a fairly accurate prediction. Since we had to predict a future event, we used the concept of Forecasting. Forecasting is a method which makes use to previous data to tell if a similar situation is expected in near future. It watches out for the changes that took place and how the result was built, hence making it an appropriate approach for the problem given above.

Link to Final Project

https://github.com/abyasingh/spaceapp_challenge

Link to Project "Demo"

https://drive.google.com/drive/folders/1ivxnch-61rB3W7oi_1t-CG7sDHu8-hP8?usp=sharing

Detailed Project Description

Language used : Python

Notebook used : Jupyter

Tools : Tensorflow, NumPy, Pandas, cdflib, xarray, spacepy

Space Agency Data

https://cdaweb.gsfc.nasa.gov/pub/data/wind/mfi/mfi_h2/2022/

https://cdaweb.gsfc.nasa.gov/pub/data/dscovr/h0/mag/2022/


We used the data from these souces to determine what exact changes took place during the occurnece and what changes we must look out for, that may indicate that another Carrington event might actually be underway.

Hackathon Journey

  1. We chose this project as we were genuinely curious as to when the next Carrington may take place and cause a major disruption in the modern world. A bit research made it clear that this event can actually cause a lot of issues and loss if it suddenly takes place in near future.
  2. We had to figure out what each heading in the data meant as the titles such as 'B1F1' didn't help. We used cdflib to get detailed information about the same.
  3. During the journey we realized the importance of sorting data according to relevance; converting the data from one form to another was also something new we learnt as earlier mostly the data was provided in default form.
  4. Merging and concatenating data caused a bit problem as the database was large and at a point of time, it consumed the entire RAM. To resolve this, we exceted the commands in some slots in ranges compatible to the system.

References

https://cdaweb.gsfc.nasa.gov/pub/data/wind/mfi/mfi_h2/2022/

https://cdaweb.gsfc.nasa.gov/pub/data/dscovr/h0/mag/2022/

Save the Earth from Another Carrington Event!

If a major space weather event like the Carrington Event of 1859 were to occur today, the impacts to society could be devastating. Your challenge is to develop a machine learning algorithm or neural network pipeline to correctly track changes in the peak solar wind speed and provide an early warning of the next potential Carrington-like event.