Dark Mode
Forest Fire Weather Prediction Dataset
Data Science and Analytics
Tags and Keywords
Trusted By
"No reviews yet"
Free
About
This dataset focuses on forest fire incidents within the Montesinho Natural Park in Portugal, providing records for 517 separate events. Its primary purpose is to enable the prediction of burnt areas resulting from these fires. Each entry includes geographical coordinates, the day and month of the incident, and the total area burnt in hectares. Furthermore, it incorporates vital meteorological variables such as rainfall, temperature, humidity, and wind speed, which are crucial in understanding fire behaviour and spread. The dataset offers a detailed view of environmental factors impacting forest fires.
Columns
- X: X-axis spatial coordinate within the Montesinho park map, ranging from 1 to 9.
- Y: Y-axis spatial coordinate within the Montesinho park map, ranging from 2 to 9.
- month: The month of the year when the fire occurred, expressed as a three-letter abbreviation (e.g., "jan" to "dec"). 'August' and 'September' are the most common months.
- day: The day of the week, expressed as a three-letter abbreviation (e.g., "mon" to "sun"). 'Sunday' is the most common day.
- FFMC: Fine Fuel Moisture Code index from the Fire Weather Index (FWI) system, with values ranging from 18.7 to 96.20.
- DMC: Duff Moisture Code index from the FWI system, with values from 1.1 to 291.3.
- DC: Drought Code index from the FWI system, with values from 7.9 to 860.6.
- ISI: Initial Spread Index from the FWI system, with values from 0.0 to 56.10.
- temp: Temperature in Celsius degrees, ranging from 2.2 to 33.30.
- RH: Relative humidity in percentage, ranging from 15.0 to 100.
- wind: Wind speed in kilometres per hour, ranging from 0.40 to 9.40.
- rain: Outside rain in millimetres per square metre, with values from 0.0 to 6.4.
- area: The burnt area of the forest in hectares, ranging from 0.00 to 1090.84.
Distribution
This dataset is provided as a CSV file (forestfires.csv), approximately 25.48 kB in size. It comprises 517 records, with each record containing 13 distinct columns. The structure is tabular, with no missing values observed across any of the columns, ensuring a clean and ready-to-use format.
Usage
This dataset is highly suitable for various analytical and modelling tasks, including:
- Predicting forest fire spread and the extent of burnt areas.
- Analysing seasonal trends and the impact of meteorological conditions on fire occurrences.
- Building regression or classification models based on spatial, temporal, and weather-related variables.
- Conducting exploratory data analysis to uncover patterns and relationships within fire incidents.
Coverage
The dataset specifically covers forest fire incidents within the geographical confines of the Montesinho Natural Park in Portugal. It spans all months of the year ("jan" to "dec") and all days of the week ("mon" to "sun"), providing a temporal scope that allows for seasonal and daily trend analysis. The data includes meteorological variables and spatial coordinates for each fire. The data is expected to be updated annually.
License
CC0: Public Domain
Who Can Use It
This dataset is ideal for:
- Data scientists and machine learning practitioners developing predictive models for environmental hazards.
- Researchers studying climate change impacts, ecological patterns, and disaster management strategies.
- Environmental agencies and forestry management organisations for risk assessment and resource allocation.
- Students and beginners in data science looking to practise exploratory data analysis, regression, and feature engineering with real-world environmental data.
Dataset Name Suggestions
- Montesinho Forest Fire Incidents
- Portuguese Forest Fire Data
- Forest Fire Weather Prediction Dataset
- Montesinho Wildfire Analytics
- European Forest Fire Dataset
Attributes
Original Data Source: Forest Fire Weather Prediction Dataset
QUALITY