Analysing Emissions from the Fires in Europe

Fires in Southern Europe

Climate change has catalyzed drastic changes in wildfire patterns across the globe resulting in increased intensity and higher frequency of wildfires. Europe is no exception to this phenomenon as wildfire risk soars with each passing season causing inevitable increases in the exposure of businesses to material financial losses across the corporate value chain.

2019 may be cited as one of the most notable evidence of this increased risk as Europe witnessed the worst-ever fires in recent history, resulting in over 400,000 hectares (ha) of the continent’s natural land burned and a record amount of nature protection zones adversely impacted.

While wildfires may be anthropogenic or naturally caused, their propensity to grow i.e. the f ire risk, may be influenced by a number of factors such as climatic conditions (e.g. humidity, temperature, and wind), vegetation (e.g. fuel load and condition), topography, forest management practices, and the socio-economic context. Hence, forest fires may have high inter-annual variability and be determined strongly by meteorological conditions.

As the EU strategy on climate change adaptation recognizes, climate change is projected to have a significant impact on forest fire risk in Europe. As a result, numerous fire control efforts have been adopted in Europe including the European Forest Fire Information System (EFFIS).

However, despite the enhanced preparedness of EU members, 340,000 hectares (ha) of land was burned in 2020, an area 30 percent larger than Luxembourg. According to official statistics, Romania was the country most hit by forest fires in 2020, followed by Portugal, Spain, and Italy.

In fact, Portugal, Spain, France, Italy, and Greece are among the top five countries in Europe in terms of the number of fires and total burnt area, over a period of 40 years, i.e.,1980 - 2020.

Burnt Area in Europe over 40 years, Source: EFFIS

Most recently, Europe witnessed an even worse fire season in 2021. And now, in 2022, the continent is experiencing a continuation of the same trend. The impact is more severely observed in Southern Europe including countries like Croatia and Romania, where the number of fires are way over the 15-year average.

Stacked bar plot showing the annual total average number of fires in EU from 2006-2021 vs total fires in 2022. Source: European Forest Fires Information System, Current Statistics

In this tutorial, we will explore how the GHG emissions from our fires dataset can be used to monitor any biomass events as well as associated emissions with a focus on France, Portugal, Spain, Greece, and Italy, i.e., in Europe’s most fire-affected countries.

💡 Before we can query anything from the BSA’s APIs, we need the shape IDs for the relevant regions. For this purpose, we will utilize the Shape Explorer tool found in SpaceTime. Shape Explorer aims to make it possible for users to use consistent shape IDs that are unique and represent a specific boundary at different administrative levels. This unique shapeId lets us filter data spatially and fetch all data falling within the shape. It will help in preventing any non-uniformity that can result from location names which may differ from person to person.

Setting up the environment and fetching the data

Fire up Jupyter Notebook in the terminal by typing jupyter notebook and installing the required packages. For more assistance on fetching the data refer to our previous tutorial here.

# Getting the fire and emissions data
params_portugal = {
    "api-key": "YOUR BSA API KEY HERE",
    "shapeId": "213e2599-2cc0-462b-9e62-392aa56868f0",
    "startDate": "2018-01-01T00:00:00.000Z",
    "endDate": "2021-12-31T00:00:00.000Z",
    "raw": "false",
    "adminLevelBucket": "2",
    "keys": "co2_emission_estimate",
}

# Making an API call
portgual_data = requests.get("https://gateway.blueskyhq.io/api/zuri/fires?",
                             params=params_portugal).json()

# Getting the data in a more accessible format -> Pandas Dataframe
portugal = pd.DataFrame(portugal_data['data'])

Similarly, we can get the fire and emissions data for the remaining 4 countries.

One key point to note here is that the shapeId is unique to all the 5 countries of interest which can be accessed from Shape Explorer. More on how to use Shape Explorer is explained in detail in our previous tutorial - Calculating Emissions from California & Oregon fires.

A combined dataframe containing all the data from the 5 countries will help us in visualizing the fire scenario in Southern Europe.

# Combining the fires data in one dataframe
south_eu_fires = pd.concat([portugal, spain, france, italy, greece])

south_eu_fires.index = pd.to_datetime(south_eu_fires.date)

# Getting a year column
south_eu_fires["year"] = pd.to_datetime(south_eu_fires["datetime"]).dt.year

The combined dataframe containing data from the 5 Southern EU countries, BSA.

Moving forward the combined dataframe can be used to make interactive visualizations and extract useful insights.

Monthly Distribution of Fires

We will now use BSA’s GHG Emissions Dataset to visualize the monthly distribution of fires from 2018-2021.

Source: BSA

For the rest of the tutorial, we will focus on visualizing how the fires and emissions from fires have evolved in the 5 countries over time. We will use Python packages - Plotly Express, Mapbox along with BSA’s GHG emissions dataset to achieve it.

# To install Plotly Express and Mapbox in the Jupter Notebook
!pip install plotly-express
!pip install mapbox

Yearly Evolution of Fires

# Resampling to get the Yearly Data 
annual_portugal_fires = south_eu_fires[south_eu_fires['States'] ==
                                       'Portugal'].resample('Y').sum()
annual_spain_fires = south_eu_fires[south_eu_fires['States'] ==
                                    'Portugal'].resample('Y').sum()
# Repeat the same for the remaining countries

# Create a new lat, lon for the 5 countries
# Centroid coordinates for the 5 countries that has been used in this tutorial
# portugal (lat, lon) : (39.754, -8.374)
# spain    (lat, lon) : (39. 585, -3.370)
# france   (lat, lon) : (46.791, 2.450)
# italy    (lat, lon) : (43.021, 12.602)
# greece   (lat, lon) : (39.558, 22.447)

annual_portugal_fires['lat'] = 39.754
annual_portugal_fires['lon'] = -8.374

# Similar steps can be followed for the remaining 4 countries
# Combining into a single dataframe using pd.concat
south_eu_fires_viz = pd.concat([
    annual_portugal_fires, annual_spain_fires, annual_france_fires,
    annual_italy_fires, annual_greece_fires
])

Now we are interested in seeing how the intensity of fires is changing. For this, we will plot Bubble Plot over a Mapbox Satellite base map.

💡 Mapbox API token can be generated for free from here.

import plotly.express as px

# Your Mapbox API token to be set using px.set_mapbox_access_token()
px.set_mapbox_access_token('YOUR MAPBOX API TOKEN HERE')

# Visualising the evolution of fires
fig = px.scatter_mapbox(south_eu_fires_viz,
                        lat="lat",
                        lon="lon",
                        color="fires",
                        size='fires',
                        size_max=75,
                        zoom=5,
                        mapbox_style="satellite-streets",
                        animation_frame='year',
                        color_continuous_scale=px.colors.sequential.YlOrRd)
fig.show()

Fires in Portugal, Spain, France, Italy and Greece for 2019 were estimated using BSA’s GHG Emissions from Fires dataset.

Fires in Portugal, Spain, France, Italy and Greece from 2018-2021 were estimated using BSA’s GHG Emissions from Fires dataset.

Yearly Evolution of GHG Emissions from Fires

The same steps can be repeated to visualize the evolution of CO2 emissions from fires.

# Your Mapbox API token to be set using px.set_mapbox_access_token()
px.set_mapbox_access_token('YOUR MAPBOX API TOKEN HERE')

# Visualising the evolution of fires
fig2 = px.scatter_mapbox(south_eu_fires_viz,
                         lat="lat",
                         lon="lon",
                         color="co2_emission_estimate",
                         size='co2_emission_estimate',
                         size_max=75,
                         zoom=5,
                         mapbox_style="satellite-streets",
                         animation_frame='year',
                         color_continuous_scale=px.colors.sequential.YlOrRd)
fig2.show()

CO2 emissions from fires in Portugal, Spain, France, Italy and Greece from 2018-2021 estimated using BSA’s GHG Emissions from Fires dataset.

We can see the yearly trend of fires and co2_emission_estimate by plotting line charts of both the parameters.

fig = px.line(south_eu_fires_viz, x="year", y="fires", color='Country', markers=True,
             title="Total Number of Fires")

fig.update_layout(
    xaxis = dict(
        tickmode = 'array',
        tickvals = [2018, 2019, 2020, 2021],
        ticktext = ['2018', '2019', '2020', '2021']
    )
)

fig.show()

# Similarly, we can plot the co2_emission_estimate chart

Key Takeaways

From the monthly distribution of the fires plot, we can conclude that the months from July till September saw the most number of fires during the 4-year period.
Spain saw 180% increase in the number of fires in 2021 from 2018 resulting in a nearly 1400% increase in CO2 emissions from fires.
Greece experienced a 170% increase in the number of fires in 2021 from 2018, Italy experienced an increase of around 100%, France had almost little to no change in the number of fires and Portugal experienced a decrease by almost 40%.
While the number of fires in France remained the same over 4 years, the CO2 emissions showed a hike of almost 147% clearly showing that the intensity of fires is increasing over time.
In terms of the number, Italy has the highest fire count in all years.