chitreshkr / Database

Created a database of US Wildfires of 188 million data using Mysql and Nosql(MongoDB)

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

US Wildfire Database Project

Database Project

Entity-Relationship Diagram

ER

Assumptions/Notes About Data Entities and Relationships

  1. Every individual fire incident can be reported from only one major source dates
  2. Every single location can have only one land owner and an owner can have lands at many locations
  3. Data time stamps are irrespective of time zones and all the recorded are in same time zone
  4. NWCG is considered as major primary organization of data source
  5. MTBS dataset is considered as secondary dataset
  6. Statistical cause of fire can be only one type
  7. One reporting agency has multiple units
  8. One fire must have one control date, but one controlled date can be used by many fires

The model is in 3NF because the functional dependency has been removed by creating different tables from fire table. Owner description is dependent on the owner code and depends on the location, so we connected location and the owner . Fire_code ,ICS Incident Number, ICS Name ,Local Incident ID ,Local Fire Report ID has Null values so we have kept it in the fire table only because we can't fetch the data for all the records.

Physical Database

Assumptions/Notes About Data Set

Include any assumptions made about data such as empty fields, sparse data, bad data, etc.

  1. Null values have been found in more than 5 columns. That is; Fire_code, ICS_209, Incident_number and Name, MTBS_ID, MTBS_FIRE_NAME, County, FIPS_CODE, FIPS_NAME.
  2. NWCG_REPORTING_UNIT_NAME and similarly, SOURCE_REPORTING_UNIT_NAME have multiple multivalued rows
  3. FIRE can be reported by only unit_ID
  4. One unit can have multiple departments
  5. One agency has multiple units
  6. Multiple agencies can be at one location
  7. One department has multiple units
  8. A fire is reported with NWCG unit

Screen shot of Physical Database objects

Physical_database

Data in the Database

Keys

About

Created a database of US Wildfires of 188 million data using Mysql and Nosql(MongoDB)