In Brief 

This dataset offers modeled daily fire danger time series, driven by seasonal weather forecasts. It provides long-range predictions of meteorological conditions conducive to the initiation, spread, and persistence of fires. The fire danger metrics included in this dataset are part of an extensive dataset produced by the Copernicus Emergency Management Service (CEMS) for the European Forest Fire Information System (EFFIS) and the Global Wildfire Information System (GWIS). EFFIS and GWIS are used for monitoring and forecasting fire danger at both European and global scales. The dataset incorporates fire danger indices from the U.S. Forest Service National Fire-Danger Rating System (NFDRS), the Canadian Forest Service Fire Weather Index Rating System (FWI), and the Australian McArthur (Mark 5) rating systems.

This dataset was generated by driving the Global ECMWF Fire Forecast (GEFF) model with seasonal weather ensemble forecasts from the European Centre for Medium-Range Weather Forecasts (ECMWF) System 5 (SEAS5) prediction system.These forecasts initially consist of 25 ensemble members until December 2016, referred to as re-forecasts. After that period, they consist of seasonal forecasts with 51 members. It is important to note that the re-forecast dataset was initialized using ERA-Interim analysis data, while forecast simulations from 2016 onward are initialized using ECMWF operational analysis. Therefore, it is suggested that the period 1981-2016 be used as a reference period, while the period 2017-to present as a real time forecast.

For both the re-forecast (1981-2016) and forecast periods (2017-present), the temporal resolution is daily forecasts at 12:00 local time, available once a month, with a prediction horizon of 216 days (equivalent to 7 months). The data records in this dataset will be extended over time as seasonal forcing data becomes available. Once the SEAS5 operation ceases, the dataset will be updated with the next ECMWF seasonal system (SYS6). It is essential to note that this is not a real-time service, as real-time forecasts are accessible through the EFFIS web services.

These seasonal forecasts can be used to assess the performance of the forecasting system or to develop tools for statistically correcting forecast errors. ECMWF produces this dataset as the computational center for fire danger forecasting within the Copernicus Emergency Management Service (CEMS) on behalf of the Joint Research Centre, which serves as the managing entity for this service.

Fire danger variables descriptions

The Canadian Fire Weather index 

Schematic of the FWI

The Canadian Fire Weather Index (FWI) is a system used in Canada to assess the potential risk and behavior of forest fires. It provides a numerical rating that indicates the relative ease of ignition and the potential intensity of fire spread in forest fuels. The FWI system incorporates various weather and fuel moisture measurements to generate indices that collectively describe the fire danger level. The purpose of the Canadian Fire Weather Index is to assist fire managers, fire behavior analysts, and meteorologists in making informed decisions regarding fire prevention, preparedness, and suppression strategies. It helps in allocating firefighting resources efficiently by identifying areas with high fire risk and potential fire behavior. There are various indices that are provided

a. Fine Fuel Moisture Code (FFMC):
The FFMC represents the moisture content of surface organic materials, such as grasses, needles, and small twigs. It quantifies the ease of ignition and the flammability of fine fuels. The FFMC ranges from 0 to 101, where higher values indicate drier and more easily ignitable fuels.

b. Duff Moisture Code (DMC):
The DMC measures the moisture content of decomposed organic material beneath the surface layer. It represents the availability of fuel for smoldering fires. The DMC ranges from 0 to 1000, with higher values indicating drier conditions.

c. Drought Code (DC):
The DC quantifies the moisture content of deep, compact organic layers. It reflects the droughtiness of deep fuels and their potential for sustaining intense, high-severity fires. The DC ranges from 0 to 1000, with higher values indicating drier conditions.

d. Initial Spread Index (ISI):
The ISI estimates the potential rate of fire spread immediately after ignition. It considers wind speed and the FFMC. The ISI ranges from 0 to 50, with higher values indicating a faster fire spread potential.

e. Buildup Index (BUI):
The BUI represents the total amount of fuel available for combustion. It considers the DMC and DC. The BUI ranges from 0 to 1000, with higher values indicating a greater quantity of available fuel.

f. Fire Weather Index (FWI):
The FWI is the composite index that summarizes the overall fire danger. It combines the FFMC, DMC, and DC into a single value. The FWI ranges from 0 to 100, with higher values indicating more severe fire weather conditions.

The Canadian Fire Weather Index values are calculated using complex mathematical equations based on observed weather data and fuel moisture codes. These calculations are typically automated using specialized software or online tools. The resulting numerical values can be interpreted as follows:

- FFMC, DMC, and DC values below 80 generally indicate low fire danger.
- FFMC, DMC, and DC values between 80 and 90 indicate moderate fire danger.
- FFMC, DMC, and DC values above 90 indicate high fire danger.
- ISI values above 10 indicate high potential for fire spread.
- BUI values above 40 indicate an increasing potential for large fires.
- FWI values above 30 indicate high fire danger conditions.

It is important to note that interpretation guidelines may vary based on regional standards and operational practices. Local fire management agencies may provide specific thresholds and guidelines for their respective areas.

The Canadian Fire Weather Index is used in a range of applications, including:

- Wildfire management and suppression: It helps fire managers assess the current and forecasted fire danger levels and allocate firefighting resources accordingly.
- Prescribed burning: The FWI assists in determining suitable conditions for conducting controlled burns, reducing the risk of unplanned wildfires.
- Fire danger rating: The FWI system aids in developing fire danger rating systems to inform the public, land managers, and emergency response agencies about the level of fire risk in a given area.

While the Canadian Fire Weather Index provides valuable information for assessing fire danger, it has certain limitations:

- It does not account for the presence of ignitions sources, such as lightning strikes or human activities.
- It does not incorporate topographic influences, which can significantly affect fire behavior.
- The accuracy of the FWI is dependent on the availability and quality of weather and fuel moisture data.

 References:
- Canadian Forest Service. (2018). Fire Weather Index System: User's Guide. Natural Resources Canada.

- Canadian Wildland Fire Information System (CWFIS). (n.d.). Fire Weather Index System.

The National Fire Danger Rating System (NFDRS) 

Schematic for the NFDRS

The National Fire Danger Rating System (NFDRS) is a comprehensive fire danger assessment and prediction system used in the United States. It provides standardized methods for evaluating fire danger conditions based on weather, fuels, and fire behavior factors. The purpose of the NFDRS is to assist fire management agencies, firefighters, and land managers in assessing fire danger, predicting fire behavior, and making informed decisions regarding fire prevention, preparedness, and suppression strategies. There are various components of the National Fire Danger Rating System:

a. Fuel Models:
The NFDRS incorporates different fuel models that represent various vegetation and fuel types across different regions. Fuel models provide information on fuel load, fuel moisture content, and fuel characteristics, which are essential for fire danger assessment and fire behavior prediction.

b. Fire Behavior Prediction (FBP) System:
The FBP system utilizes mathematical equations and empirical models to estimate fire behavior parameters, such as rate of spread, flame length, and fireline intensity. It considers weather conditions, fuel moisture, and fuel models to predict fire behavior under different scenarios.

c. Fire Danger Rating (FDR) System:
The FDR system combines weather, fuel moisture, and fire behavior factors to generate fire danger ratings. It provides standardized indices that indicate the level of fire danger and potential fire behavior. The FDR system helps fire managers assess and communicate the fire danger to personnel and the public.

d. Fire Danger Indices:
The NFDRS calculates several fire danger indices, including the Energy Release Component (ERC), Burning Index (BI), Spread Component (SC), and others. These indices represent different aspects of fire danger, such as fuel flammability, potential rate of fire spread, and potential fire intensity.

The National Fire Danger Rating System utilizes complex calculations based on observed weather data, fuel models, and fire behavior equations. The resulting numerical values and indices can be interpreted as follows:

- Fire danger indices are typically rated on a numerical scale, such as a 1-100 or a 0-10 scale.
- Higher values indicate more severe fire danger, increased potential for fire spread, and greater fire behavior intensity.
- Interpretation guidelines and thresholds may vary based on regional standards, local practices, and fire management agency guidelines.

Fire danger ratings often include descriptors, such as low, moderate, high, very high, and extreme, to provide additional context for interpreting the numerical values.

The NFDRS has various applications, including:

- Fire management and suppression: It assists fire managers in assessing fire danger levels, predicting fire behavior, and allocating firefighting resources effectively.
- Prescribed burning: The system helps in planning and implementing controlled burns by considering fire danger and potential fire behavior.
- Fire weather forecasting: The NFDRS aids meteorologists in providing fire weather forecasts and warnings to support fire management operations.

While the National Fire Danger Rating System is a valuable tool, it has certain limitations:

- It relies on accurate and up-to-date weather data and fuel moisture measurements.
- The system may not fully account for local variations in topography and vegetation characteristics.
- Fire behavior can be influenced by factors not explicitly considered in the system, such as the presence of structures or human-made modifications.

 References:
- National Wildfire Coordinating Group (NWCG). (2004). National Fire Danger Rating System: 2016.

- Andrews, P. L. (1986). BEHAVE: Fire Behavior Prediction and Fuel Modeling System—FUEL Subsystem. Gen. Tech. Rep. INT-194. Ogden, UT: U.S. Department of Agriculture, Forest Service, Intermountain Research Station.

McArthur's Fire Danger Rating System

The first MRk4 meter

McArthur's Fire Danger Rating System is a widely used fire danger assessment system in Australia. It provides a quantitative rating of fire danger based on weather and fuel moisture conditions. The system was developed by Dr. Richard McArthur in the 1960s and has undergone several revisions since then.The purpose of McArthur's Fire Danger Rating System is to assess and communicate the potential risk and behavior of bushfires and wildfires. It helps fire agencies, land managers, and the public make informed decisions regarding fire prevention, preparedness, and response strategies. There are several components of McArthur's Fire Danger Rating System:

a. Fire Danger Index (FDI):
The Fire Danger Index represents the overall fire danger rating. It considers weather variables such as temperature, relative humidity, wind speed, and rainfall, as well as fuel moisture content. The FDI ranges from 1 to 100, with higher values indicating more severe fire weather conditions and increased fire danger.

b. Grassland Fire Danger Index (GFDI) (not provided):
The Grassland Fire Danger Index specifically assesses fire danger in grassland areas. It takes into account weather conditions and fuel moisture content relevant to grassland fuels. The GFDI ranges from 1 to 100, with higher values indicating higher fire danger in grassland environments.

c. Forest Fire Danger Index (FFDI) (not provided):
The Forest Fire Danger Index is designed to evaluate fire danger in forested areas. It considers weather conditions and fuel moisture content specific to forest fuels. The FFDI ranges from 1 to 100, with higher values indicating higher fire danger in forested environments. 

McArthur's Fire Danger Rating System uses mathematical equations to calculate the Fire Danger Index (FDI), Grassland Fire Danger Index (GFDI), and Forest Fire Danger Index (FFDI). These calculations incorporate observed weather data, such as temperature, humidity, wind speed, and rainfall, along with fuel moisture information. The resulting numerical values can be interpreted as follows:

- FDI, GFDI, and FFDI values below 12 generally indicate low fire danger.
- FDI, GFDI, and FFDI values between 12 and 24 indicate moderate fire danger.
- FDI, GFDI, and FFDI values above 24 indicate high fire danger.
- Higher values within each range indicate increasing severity of fire danger.

It is important to note that specific interpretations and thresholds may vary based on regional standards, local practices, and fire management agency guidelines.

McArthur's Fire Danger Rating System has various applications, including:

- Fire management and suppression: It assists fire agencies in assessing fire danger levels, predicting fire behavior, and allocating firefighting resources effectively.
- Fire permits and restrictions: The rating system helps determine the need for fire permits and implement fire restrictions based on the assessed fire danger.
- Public awareness and education: The system aids in communicating fire danger to the public and promoting awareness of fire safety measures during periods of elevated fire risk.

While McArthur's Fire Danger Rating System is a valuable tool, it has certain limitations:

- It does not consider local variations in topography, which can significantly influence fire behavior.
- The accuracy of the system relies on the availability and quality of weather data and fuel moisture measurements.
- The rating system does not account for local fire history or specific fire-prone vegetation types.

References:
- McArthur, A. G. (1973). Forest Fire Danger Meter. Australian Forestry, 36(1), 39-45.
- McArthur, A. G., Noble, I. R., & Bary, G. A. V. (1982). Forest fire danger meter-1977. Forest and Timber Bureau, Canberra, Australia.

- Bureau of Meteorology (Australia). (n.d.). Fire Danger Rating and the McArthur Forest Fire Danger Index. 

Available variables

We provide the following subset of variables 

Time Interpolation to local noon

A model integration at any nominal time will simulate the atmospheric conditions at a different local time depending on the location. A temporal and spatial collage of 24-h time model simulations is performed to produce a snapshot at 1200 local time. Thus, temperature and relative humidity fields are cut into, for example, 3-hourly time strips using the closest 3-h forecast output and then concatenated together so that the final field is representative of the conditions around the local noon within the 3-h resolution available. Using this method, the driving forcing are a composite of forecast outputs at different lead times in a 24-h interval and could therefore have different forecast accuracy. This inconsistency is assumed insignificant given the limited difference in forecast skills in a 24-h lead time range (Buizza et al. 1999)^[1]. The most common way to achieve a 12 local time field is through a concatenation of fields. Adjacent stripes of forecasts can be sliced together as highlighted in Figure. The simplicity of the approach however implies that artefact can be introduced at the interface between two slices.

Concatenation of forecasts to derive a field at 12 local time everywhere. The discontinuity line represents the change of date. The stripes are taken from the forecast times specified at the bottom.

ECMWF has developed a new interpolation method performing a weighted average between the two closest timesteps. Below is a graphical explanation of the way the two methods are implemented is provided. If we extract the time series of two locations 1 and 2 near a change of time, the 24 hours forecast will provide the diurnal time for temperature (here chosen as an example) in those points. If we are interested in the value of temperature at 12 local time this will be the value of the prediction at 12 UTC for point 1 and 15 UTC for point 2 in the assumption of a 3-hour resolution forecast. The choice of the different forecast will create a discontinuity in the fields as depicted in the map. The new method instead will interpolate for both points between the value at 12 UTC and the value at 15UTC by weighting the two temperatures by their closeness to any available forecast. This new method provides a much closer agreement to the real diurnal cycle and removes the boundary artefacts that were presented above.

Usage notes 

Download global fire danger forecast maps

Downloading these data is rather straight forward using the CDS web interface. The registered user needs to tick a few boxes to specify the index, period of interest and type of data, then click on a `Download' button. For larger data requests, the use of the CDS API is recommended. Below an example script is provided.

import cdsapi

c = cdsapi.Client()

c.retrieve(
    'cems-fire-seasonal-reforecast',
    {
        'format': 'grib',
        'variable': 'fire_weather_index',
        'system_version': '4_1',
        'year': '1985',
        'month': '09',
        'leadtime_hour': [
            '12', '36', '60',
            '84', '108', '132',
            '156', '180', '204',
            '228', '252', '276',
            '300', '324', '348',
            '372', '396', '420',
            '444', '468', '492',
            '516', '540', '564',
            '588', '612', '636',
            '660', '684', '708',
        ],
    },
    'download.grib')

Plotting data using the CDS toolbox

To harness the power of the CDS, users are invited to familiarise with the CDS Toolbox. This is an interactive environment that allows to process and plot data without necessarily downloading them. This is particularly useful for users with limited bandwidth and/or unstable connections. The toolbox is designed to develop python applications that can be shared with other users, hence streamlining collaborative research and development. The script below can be pasted in the toolbox editor to generate a static map of the Fire Weather Index (as they are shown in the EFFIS and GWIS platform) that can be exported and used for reports and publications. 

import cdstoolbox as ct
# Magics plot configuration dictionary
MAP_CONFIG = {    
    'contour': {
        'contour_level_selection_type': 'level_list',
        'contour_level_list': [0, 5.2, 11.2, 21.3, 38, 50, 150],
        'contour_shade': 'on',
        'contour_label': 'off',
        'contour_shade_method': 'area_fill',
        'contour_shade_colour_method': 'list',
        'contour_shade_colour_list': ['#84F07F', '#FFEB3C', '#FFB00C',
                                      '#FA4F00', '#B40000', '#280923'],
        'contour': 'off',
        'legend': 'on',
    },
    'legend': {
        'legend_text_colour': 'black',
        'legend_text_font_size': 0.4,
        'legend_display_type':'continuous',
    }
}
# Initialise the application
@ct.application(title='Fire Weather Index 2020-07-01', fullscreen=True)
@ct.output.figure()
def application():
    # Retrieve full resolution FWI data for a single date
    data = ct.catalogue.retrieve(
        'cems-fire-historical-v1',
        {
            'product_type': 'reanalysis',
            'variable': 'fire_weather_index',
            'dataset_type': 'consolidated_dataset',
            'system_version': '4_1',
            'year': '2020',
            'month': '07',
            'day': '01',
            'grid': '0.5/0.5',
        }
    )
    # Plot the data using the defined configuration MAP_CONFIG on a dynamic map
    plot = ct.map.plot(data, **MAP_CONFIG)
     
    return plot