Applying a Clustering Model to Forest Fires

In one of my recent graduate degree courses I was required to look at current events for which a clustering model would be appropriate. I needed to list up to five predictors that we might use.

A situation from current events is that recently and in the past few years there have been a number ofwildfires and forest fires. I decided to do some small web research to figure out whether the cause of wildfires, forest fires or other types of fires would be appropriate for a clustering model. I believe that yes, a clustering model would be appropriate.

For example you could use a clustering model to cluster the types of fires and their origins, whether they be fires started by humans, fires started by nature (e.g. lighting, combustion) and you could also cluster their subtypes within human caused fires, versus natural causes. You could also cluster fires by many other factors such as climate, deforestation, agricultural impacts, the list can go on. For my assignment I picked the following predictors.
5 predictors that we might use:

• Fine Fuel Moisture Code – The Fine Fuel Moisture Code (FFMC) is a numerical rating of the moisture content of litter and other cured fine fuels. This code is an indicator of the relative ease of ignition and flammability of fine fuel.
• Duff Moisture Code – The Duff Moisture Code (DMC) is a numerical rating of the average moisture content of loosely compacted organic layers of moderate depth. This code gives an indication of fuel consumption in moderate duff layers and medium-size woody material.
• Drought Code – The Drought Code (DC) is a numerical rating of the average moisture content of deep, compact, organic layers. This code is a useful indicator of seasonal drought effects on forest fuels, and amount of smouldering in deep duff layers and large logs.
• Grass Fuel Moisture (GFM)– The GFM is a fourth fuel moisture category for grass fuel moisture specifically (Wotton, 2009). Research in Ontario (Kidnie et.al, 2010) quantified the fuel moisture trends for grass fuels and established a grass fuel moisture model that is produced only with hourly data.
• Cause of Fire – Human, vs Natural Causes (e.g. Lightning, Extreme Heat/Combustion) – Human causes include – A glass bottle thrown on the side of the road magnifies a ray of sunlight, igniting grass around it. Kids playing with matches inadvertently start a fire. A campfire is not properly extinguished and spreads. Sparks from a train ignite grass around the tracks, etcetera.
• References
  Author Not Identified. Date Published Not Shown. Boreal Forests of the World. The Canadian Forest Fire Weather Index (FWI) System. Retrieved from URL
  http://www.borealforest.org/world/innova/fire_prediction.htm
  Author Not Identified. (2019, March 15). National Wildfire Coordinating Group. Fire Weather Index (FWI) System. Retrieved from URL
  https://www.nwcg.gov/publications/pms437/cffdrs/fire-weather-index-system#TOC-FWI-Fuel-Moisture-Codes
  Bradford, A. (2018, August 29). Wildfires: Causes, Costs & Containment. Live Science. Retrieved from URL
  https://www.livescience.com/63458-wildfires.html

https://www.bigendiandata.com/2017-10-24-Forest_Fires/

https://www.kaggle.com/johndoea/kmeans-forest-fire-clustering

I update my website with my laptop as found on Amazon: https://amzn.to/3bXi4W7 Note: As an Amazon Associate I earn from qualifying purchases.