Forest Fires in Brazil¶

Number of forest fires reported in Brazil divided by states¶

Luís Gustavo Modelli - Kaggle updated 6 months ago

Tags¶

business, sensitive subjects, agriculture and forestry, fire and security services

Description¶

Forest fires are a serious problem for the preservation of the Tropical Forests. Understanding the frequency of forest fires in a time series can help to take action to prevent them. Brazil has the largest rainforest on the planet that is the Amazon rainforest.

Content¶

This dataset report of the number of forest fires in Brazil divided by states. The series comprises the period of approximately 20 years (1998 to 2017). The data were obtained from the official website of the Brazilian government.

http://dados.gov.br/dataset/sistema-nacional-de-informacoes-florestais-snif

Acknowledgements¶

We thank the brazilian system of forest information

Inspiration¶

With this data, it is possible to assess the evolution of fires over the years as well as the regions where they were concentrated. The legal Amazon comprises the states of Acre, Amapá, Pará, Amazonas, Rondonia, Roraima, and part of Mato Grosso, Tocantins, and Maranhão. Data (255 KB) Data Sources amazon.csv 5 columns

Columns¶

• year
  • Year when Forest Fires happen
• state
  • Brazilian State
    • It turned out during the editing of the data that there are 3 state names starting with 'Rio': Rio Grande do Sul, Rio Grande do Norte, and Rio de Janeiro...
    • As this dataset only mentions 'Rio', I'll assume that it concerns combined data of Rio Grande do Sul and Rio Grande do Norte.
• month
  • Month when Forest Fires happen
• number
  • Number of Forest Fires reported
• date
  • Date when Forest Fires where reported

Clarification¶

• This data consists of the number of forest fires that were "reported".
  • perhaps data not retrieved from satellite images
• The number of heat spots that were detected by satellites above Brazil is higher than the forest fires numbers.
  • A heat spot indicates the existence of a fire in an element of image resolution (pixel) that varies from 1 km x1 km up to 5 km x 4 km.
  • Moreover, if the fire is too large, it may light other pixels, meaning that several heat spots will indicate one large fire.

the dataset has only data set by year, but there is an extra month column provided...

A beautifully colored graph, though not very insightful.

The number of forest fires for all the states added together¶

It appears that the numbers have been split up into monthly data over the period 1998 - 2017.

The number of forest fires for each state over the period 1998 - 2017.¶

Converting month to decimal number, so we can compare the impact of a particular month statewise.

maand2num¶

extracting the state names for retrieving the local latitude and longitude info¶

The number of forest fires for the state Acre over the period 1998 - 2017¶

The number of forest fires for the state Mato Grosso over the period 1998 - 2017¶

The states Mato Grosso, Paraiba, and Rio need a groupby operation to sum up their monthly numbers.

Build a new dataframe with the states as columns

No need for adding all columns, only the "number" columns have to be "inner" joined on the MatoGrosso index ...

MP.columns

Timeseries per state - version 1¶

make a dataframe for the DMS coordinates¶

We have to convert the DMS coordinates to decimal coordinates.¶

dfObj['z'] = dfObj['z'].apply(np.square)

DMScoordinatesBrazil.loc[2, 'Area'] = 142828.50

Convert coordinates in DMS notation to decimal coordinates¶

Plotting the reporting states of Brazil with decimal coordinates¶

selecting and plotting only the states with forest fires based per year¶

I'll make a sum of the forest fires by state and by year, and insert this data in the geodataframes.

The numbers are still too dense to visualize properly, so I started pulling out the data for each state separately. And then reconstruct it back together ´´statewise´´.
Later I fetched the unspoiled data from the Brazilian site directly.

GeoPandas maps¶

Localize the states by generating a map plot from a geopandas point geodataframe

• The place to find some data about administrative boundaries of Brazil : ftp://geoftp.ibge.gov.br/
  • The CRS adopted in IBGE's shapefiles is SIRGAS 2000 (EPSG:4674).
  • converting to wgs1984
• setting up a dataframe point and verify that it correctly interpretes the Coordinate Reference System (CRS)
• Load the boundaries and the point data, and plot them to see the locations of the states and the boundaries pattern...

Bringing the size of the state area into the picture¶

The total area of Brazil is 8,515,767 km²

Merging 2 dataframes column with area values to index value with the corresponding name of coordinates dataframe

Select only numbers from 2017, so we can compare the amount of fires to the area of the state in order to get an idea of the density of the forest fires.

The density of forest fires of each state compared to the state with maximum density in 2017.¶

Fire densities according to the Area of the state (km²)¶

Kde plot Latitude vs Longitude¶

Summary of a pivot table which splits up the fire numbers by year and by state¶

State of Tocantins, Brazil is located at latitude -10.17528 and longitude -48.2982474, 10°10'31'' S 48°17'53.7'' W. Ceara = -5.498398055555556,-39.3206238

Plot of the forest fire densities versus the state with the maximum density for 2017¶

Fetching the original data from the Brazilian site¶

The number of forest fires for the reported states over the period 1998 - 2017¶

The number of forest fires for all states together¶

The number of forest fires per state¶

I need to recombine the year and month columns to a datetime format

Forest Fires: yearly resample with mean and max.¶

Forest Fires: monthly resample with mean, standard deviation and max.¶

Total number of forest fires per year.¶

Monthly numbers per year in period 1998 - 2007¶

Monthly numbers per year in period 2007 - 2017¶

Pivot table fire numbers each state vs. year¶

Heatmap of number of forest fires split up by year and state¶

Pivot table monthly numbers vs. state.¶

Timeseries per state 1998-2017 - version 2¶

Heatmap of number of forest fires split up by month and state¶

Types of protected area¶

Protected areas, also called conservation units, are divided into different categories according to their goals. These are defined by Law No. 9.985 of 18 July 2000, which established the National System of Conservation Units (SNUC).
Objectives include conservation of nature, sustainable development, scientific research, education and eco-tourism.
Fully protected units are expected to maintain the natural ecosystem without human interference.
Sustainable use units allow sustained use of renewable environmental resources while maintaining biodiversity and other ecological attributes.

The Chico Mendes Institute for Biodiversity Conservation, which administers Federal units, defines the classes of unit as:

Fully protected (proteção integral)¶

• Ecological stations (Portuguese: Estações Ecológicas)
• Biological reserves (Portuguese: Reservas Biológicas)
• National parks (Portuguese: Parques nacionais), State parks (Portuguese: Parques estaduais) and Municipal nature parks (Portuguese: Parques naturais municipais)
• Natural monuments (Portuguese: Monumentos Naturais)
• Wildlife refuges (Portuguese: Refúgios de Vida Silvestre)

The sustainable use units are:¶

• Environmental protection areas (Portuguese: Áreas de Proteção Ambiental)
• Areas of relevant ecological interest (Portuguese: Áreas de Relevante Interesse Ecológico)
• National forests (Portuguese: Florestas Nacionais) and State forests (Portuguese: Florestas Estaduais)
• Extractive reserves (Portuguese: Reservas Extrativistas)
• Wildlife reserves (Portuguese: Reservas de Fauna)
• Sustainable development reserves (Portuguese: Reservas de Desenvolvimento Sustentável)
• Private natural heritage reserves (Portuguese: Reservas Particular do Patrimônio Natural)

In addition, some states designate areas as ecological reserve (Portuguese: Reserva Ecológica).