Important Links

Link to deployed website. Link to main project repository.

Information

This is a dummy repository used for redirecting to the original project. Please follow the link above to see the deployed website and complete repository with the necassary code.

Analysing_US_Land_Fires

The project included analysis and visualisation of US Land-fire dataset between 2013-2017.

WHAT WE DID

Steps in Data Extraction, Transformation and Load:

• Load raw data into PostgreSql
• Clean data in Python and SQL
• Create Rest APIs which gives results based on Causes, Locations, State, Year

HOW WE VISUALISED THE DATA

• Created interactive dashboard using Index html and Bootstrap CSS
• Choropleth by State with a color gradient by count of fires
• Plotly Pie chart: Filter by State and Year
• Data table using Plotly
• GeoMap with a marker cluster group layer: Filter by Year
• Plotly Charts: Bar, Bubble and Stacked Bar

WHAT WE FOUND OUT

• In the states most affected by fires, humans are the major contributor with 68%, followed by Undetermined 19% and Natural 13%
• The West coast has significantly higher counts of fires than the East coast which could be a result of drought in certain areas
• Top 5 states with high count of fires: Arizona, California, Montana, Oregon and Minnesota
• New Hampshire and Vermont did not have any reported fires and hence are excluded from the dataset
• Fires tend to be the highest in the summer months
• 2013 recorded the highest count of land fires