Fire is a key disturbance factor for the dynamics (Thonicke et al., 2001; Pausas and Keeley, 2009) and distribution (Bond et al., 2005) of the vegetation ecosystems globally. Wildland fires affect forests' function (Bowman et al., 2009), structure (Koutsias and Karteris, 2003), and adaptation (Pausas and Keeley, 2009), while significantly contributing to emissions of greenhouse gases (van Der Werf et al., 2017; Zheng et al., 2021), soil erosion (Shakesby, 2011), water and air pollution (Smith et al., 2011; Duc et al., 2018), and land cover change (van Wees et al., 2021). Wildland fires also threaten human lives and properties and can cause important socioeconomic impacts (Bowman et al., 2017, 2020).

Estimations based on coarse-resolution satellite observations indicate that around 4 Mkm2 (million kilometres squared) are burnt globally every year (Giglio et al., 2018; Lizundia-Loiola et al., 2020), although this evaluation is very conservative, as they are based on coarse-resolution satellite data, which have shown to include significant omission errors (Boschetti et al., 2019; Franquesa et al., 2022). The European territory is highly affected by wildland fires, which cause environmental, societal, and economical losses (San-Miguel-Ayanz et al., 2020, 2021). In 2021, about 500 000 ha were burnt in the European Union, of which 20 % affected Natura 2000 and other protected sites, especially in southern Europe. August was the worst month, including very large fires. Around 28 % of the total burnt area affected forest, and 25 % belonged to agricultural land types (San-Miguel-Ayanz et al., 2022). In addition, global climate change will likely increase the wildland fire risk and impacts in most of the European territory (Jones et al., 2022; IPCC, 2022). This justifies the necessity of improving the current efforts to prevent and contain wildland fires in Europe (San-Miguel-Ayanz et al., 2021).

As it is well known, the fire environment concept defines the three key elements influencing fire initiation, propagation, and effects. These are weather, topography, and fuel (Countryman, 1972). Fire behaviour is highly dependent on fuel (vegetation) characteristics, which is the only variable that can be managed to reduce fire propagation. In addition, fuel properties play a critical role in fire ignition (Alvarado et al., 2020) and in the atmospheric emissions derived from fires, particularly in the smouldering–flaming ratio of fire behaviour (Zheng et al., 2021).

Vegetation types with similar fire behaviour are grouped into fuel types and models (Pyne, 1984). The former indicate categories of vegetation with similar characteristics from a fire behaviour perspective. The latter refer to the specific parameters required to model their fire behaviour (height, load, bulk density, and particle size, among others). Fuel types refer to the surface or canopy layers. Forest understorey and low vegetation formations are surface fuels, while forest crowns and tall shrubs represent canopy fuels. Fire usually starts in surface fuels but may transfer to canopy fuels, causing crown fires, which are more dangerous than surface fires as they release more energy and propagate in larger fronts, making them harder to control (Scott and Reinhardt, 2001).

Fuel type mapping is essential in fire risk prevention, planning, and real-time fire management across multiple spatial scales (Keane et al., 2001) because it allows us to spatially describe a key factor in fire management (Keane and Reeves, 2012), while fire scientists require accurate and updated fuel maps to support fire strategic planning within a comprehensive fire danger assessment system. However, fuel mapping is challenging due to the high temporal and spatial variability in fuels (Keane et al., 2001).

The starting point of fuel mapping is the definition of a consistent fuel classification system, which includes fuel types and models (parameters). Many fuel classification systems have been developed, although the most common refer just to surface fuels (Arroyo et al., 2008), thus limiting their capability to prevent and manage crown fires (the most severe).

This is the case of the Northern Forest Fire Laboratory (NFFL) system (Anderson, 1982) and the Fire Behavior Fuel Models (FBFMs; Scott and Burgan, 2005), which have been created for the United States. Other commonly used fuel classification systems are the Fuel Characteristic Classification System (FCCS; Ottmar et al., 2007), the Canadian Fire Behaviour Prediction System (Forestry Canada Fire Danger Group, 1992), and the Mediterranean–European Prometheus system (European Commission, 1999; Arroyo et al., 2008). Although they have been developed for specific regions and conditions, they have been widely used to map fuel types in other regions (García et al., 2011; Palaiologou et al., 2013; Marino et al., 2016; Aragoneses and Chuvieco, 2021).

Fuel types have been usually mapped through fieldwork, aerial photointerpretation, ecological modelling, and existing datasets and/or remote sensing (Arroyo et al., 2008). Remote sensing methods previously applied to fuel type mapping include a wide range of techniques and input data, from medium (Palaiologou et al., 2013; Alonso-Benito et al., 2013; Marino et al., 2016; Aragoneses and Chuvieco, 2021) to high spatial resolutions (Arroyo et al., 2006; Mallinis et al., 2008). Both passive (Alonso-Benito et al., 2013; Aragoneses and Chuvieco, 2021) and active sensors (Riaño et al., 2003; González-Olabarria et al., 2012) have been used, in addition to a combination of sensors (Mutlu et al., 2008; García et al., 2011; Palaiologou et al., 2013; Marino et al., 2016).

Fuel maps exist for continental scales, such as South America (Pettinari et al., 2014) and Africa (Pettinari and Chuvieco, 2015), and global scales but include categories that are too coarse to be operationally applicable to European conditions (Pettinari and Chuvieco, 2016). However, in Europe, fuel mapping has been mostly developed for local and regional scales (Roulet, 2000; García et al., 2011; Stefanidou et al., 2020). The only European-level fuel cartography is the 2000 EFFIS (2017) fuel map (European Forest Fire Information System), which is based on land cover and vegetation maps and using the NFFL system. Other works have mapped FBFM fuel models (Scott and Burgan, 2005) for the European subcontinental scale, such as the Iberian Peninsula (Aragoneses and Chuvieco, 2021).

The lack of an adapted-to-Europe fuel classification strategy is limiting, since fuel models are site specific and should be applied to the region for which they were developed to obtain the most realistic fuel mapping and modelling (Arroyo et al., 2008). In this context, the ArcFUEL project (Bonazountas et al., 2014) proposed a methodology to enable consistent fuel mapping production over Europe, based on a hierarchical vegetation fuel classification system adapted to European conditions (Toukiloglou et al., 2013). Nevertheless, a European fuel map was not generated; only southern European countries at national (Portugal and Greece) and regional (Spain and Italy) scales were mapped (Bonazountas et al., 2014).

Considering the current limitations of European fuel mapping, this paper had three objectives. The first one was generating a fuel classification system to facilitate the integration of continental wildfire risk assessment, including both surface and canopy fuel types. The proposed classification system should be hierarchical to facilitate the integration of fuel maps at different spatial scales, include both surface and canopy fuel types, and be suitable for different purposes, from fire behaviour simulation to fire emissions or fire danger assessment. The second objective was to develop a European fuel map at 1 km spatial resolution, following the proposed fuel classification system. We aimed to develop a methodology that, by combining expert knowledge, a geographic information system (GIS), available datasets, and bioclimatic modelling, might be easily replicable and updated with low time and economic costs. Finally, the third objective was to assign surface fuel parameters to the derived fuel types, by relating them to existing fuel models. We chose the FBFM standard fuel models (Scott and Burgan, 2005), as this system is widely used and very flexible. These three objectives serve to organise the structure of this paper around three sections (Fig. 1). This work is expected to lay the framework for an integrated and homogeneous fire management strategy across European countries. The present study is part of the FirEUrisk project, which aims to create a European integrated strategy for fire danger assessment, reduction, and adaptation.

We developed the FirEUrisk hierarchical fuel classification system with the following three main requirements: it should be adapted to the great variety of European environmental conditions, include both surface and canopy fuel types, and be suitable to work at different spatial scales. The main driver of the classification system was fire behaviour modelling, but its use for fire risk assessment and fire emission estimations was also considered. To define each of the fuel types, the land cover and vegetation descriptions of the Copernicus Global Land Cover map categories (Tsendbazar et al., 2020), the UN LCCS (United Nations Land Cover Classification System) from the UNESCO (United Nations Educational, Scientific and Cultural Organization; UNESCO, 1973), the FAO (Food and Agriculture Organization, 2000), and the European Environmental Agency's CORINE Land Cover nomenclature (CLC; Kosztra et al., 2019) were used. For the wet and peat/semi-peat land fuel types, the definitions provided by the International Peatland Society (2021) were also taken into account.

The proposed hierarchical fuel classification system, FirEUrisk (Table A1 in Appendix A), encompassed a total of 85 fuel types for surface and canopy fuels, which were aggregated into six main fuel type categories, referred to the main fuel cover, which recall traditional land cover types, plus a nonfuel category. The FirEUrisk fuel classification system used several criteria to discriminate between fuel types. Subcategories were included to better estimate fuel models for each resulting fuel type category and would also lead to different fire behaviour. The first-level main categories were defined as follows:

Forest, which includes areas with tree canopy cover above 15 %, with a mean tree height ≥2 m, following the Copernicus Global Land Cover legend (Tsendbazar et al., 2020), which is based on the UN LCCS (United Nations Land Cover Classification System) from the UNESCO (United Nations Educational, Scientific and Cultural Organization; UNESCO, 1973) and the FAO (Food and Agriculture Organization, 2000). Understorey type refers to the fuel type with which the surface fire will spread in the forest.

Forest categories were divided into two vertical strata, for which the first level referred to the overstorey (canopy) characteristics and the second level to the understorey characteristics. Further subdivisions were included in the first level by considering the leaf type (broadleaf or needleleaf), the leaf deciduousness (evergreen or deciduous), and the fractional cover (open or closed). The lower stratum referred to the understorey characteristics by identifying the type of surface vegetation (grassland, shrubland, or timber litter) and its height. This allowed us to define the surface and canopy characteristics of the fuels in the forest, which can help to account for both surface and crown fires. For the rest of the main fuel types, only one vertical stratum (first level) was identified. For shrubland and grassland fuel types, subcategories were created based on fuel bed depth (height of the surface fuel layer).

Discriminating all the proposed categories may be quite challenging and should be adapted to the working scale of the fuel type product and, accordingly, to the quality of the input data available to produce it. The fuel type categories of the first level (Table 1) should be more suitable for continental or global fuel products, while the second level should be better adapted to local or regional studies, where more detailed information can be available. In this paper, the European fuel map was generated for the first level of the proposed fuel classification system, covering all European continental countries at 1 km spatial resolution. This product was developed to help the strategic planning of fire management in Europe through generating a continental map with a homogeneous and integrated fuel classification system for all countries, which would allow us to carry out standardised fire risk analysis and inform fire managers and policy-makers from a risk-wise holistic perspective for Europe.

The proposed FirEUrisk hierarchical fuel classification system was designed to be adapted to a wide range of environmental conditions, including those found in the European territory, describing both surface and canopy fuels. In this paper, we present a first product based on this classification, covering the whole European territory for the first level of the classification. We did not consider the forest understorey, the second level of the classification, which is better suited to regional and local scales where more detailed information, particularly lidar data, can be available. In any case, the obtained results constitute an improvement in European fuel mapping compared to existing fuel maps covering the European territory. The map provides more detailed categories than those of existing global fuel maps (Pettinari and Chuvieco, 2016) or the 2000 EFFIS fuel map (European Forest Fire Information System, 2017), which only referred to surface fuels and did not consider forest canopy characteristics. In addition, the FirEUrisk fuel map includes new categories such as wet and peat/semi-peat land fuel types, which are key to understanding fire emissions, and urban fuel types, crucial to prevent fire affecting humans, which were not considered in previous continental and global fuel maps.

The hierarchical nature of the system aims to define a common fuel type classification for different scales and study areas. It also offers high versatility, as it enables mapping fuels with different disaggregation of categories, depending on the detail and quality of the input data, while allowing us to overlap fuel maps for the same area at different scales, which would help the integration and comparison of fuel maps because of the common legend. Thus, whereas the fuel map developed at the European scale was based on existing European and global datasets integrated into a GIS framework, the same classification scheme could be applied to provide a more comprehensive fuel classification using a multi-sensor approach in a machine learning framework (García et al., 2011; Marino et al., 2016; Domingo et al., 2020). Its structure has similarities (e.g. hierarchical scheme) with the ArcFUEL classification (Toukiloglou et al., 2013), although this was only prepared for southern European conditions. In addition, the involvement of expert knowledge in the development of the FirEUrisk hierarchical fuel classification system suggests high acceptance, and therefore usage, among the fire risk management community in the foreseeable future. It also allowed the development of a useful classification, intended to fill the actual gaps of the European fuel mapping, towards a homogeneous and integrated fire risk prevention strategy. Nevertheless, it must be considered that the grouping of vegetation types into fuel types is a balance between the generalisation of the landscape reality and loss of detailed information, which may not be the most suitable system for all study areas.

The predicted increase in fire intensity and the occurrence of the so-called mega-fires (San-Miguel-Ayanz et al., 2013), which usually evolve from surface to crown fires, makes it necessary to improve our information on canopy fuels. Assessing the potential transition from surface to crown fires is key to prevent crown fires. For this reason, our classification approach includes both surface and canopy fuel types for the forest fuel types. Crown fires are highly influenced by the characteristics of understorey and ladder fuels, and by wildfire intensity (e.g. flame length), which is information that is not available at the European scale. However, we encourage others to complement the proposed fuel types with additional data for the regions where data may be available. This would require determining the vertical continuity of fuels, in addition to identifying the existence (or not) of a gap between the understorey and the canopy fuels strata. This might be subject of future work. The rest of the fuel types are disaggregated based on their fuel bed depth, with the thresholds suggested by experts. However, fuel mapping is still a challenge because of the high spatiotemporal variability in fuels and the need to generalise the great variety of vegetation conditions related to fire behaviour.

Regarding European fuel mapping, the combination of existing land cover and biogeographic datasets and bioclimatic models facilitated the generation of the fuel type dataset, with some of these data being specifically developed for the European conditions (i.e. Europe's biogeographic regions map, the CLC map). Nevertheless, the input datasets are a generalisation of the complex reality, with their own uncertainties and errors, which are transferred to the final European fuel map. In fact, the errors in the final fuel type dataset are similar to or even lower than those found in the main input land cover map used to obtain the fuel categories.

Estimating shrubland and grassland fuel bed depth was challenging. To the best of our knowledge, there are no large-scale, reliable datasets in Europe on these variables, which is limiting to our purposes. However, despite the models chosen to estimate surface fuel bed depth not being specifically developed for European areas, the biogeographical similarity of the regions for which they were developed, compared to European conditions, make them acceptable for our purposes. Almost all of the shrubland fuels belong to the arid/semi-arid regime, which justifies the selection of a bioclimatic model developed for an arid/semi-arid area. To avoid unrealistic estimations, we constrained the outputs to the range 0–6 m for the shrublands and to >0 m for the grasslands, while no maximum cut-off threshold was applied to the grassland category, as the obtained maximum value (1 m) was considered reasonable. In addition, the distribution of shrubland and grassland pixels led to considering the bioclimatic models to be adequate. The histogram for shrubland fuel bed depth showed the continuity of the input variable (precipitation). The histogram for the grassland fuel bed depth had an aggregated structure due to the input productivity data by biogeographic region. Obviously, the direct measurement of the shrubland or grassland fuel bed depth would be more desirable. In this sense, airborne lidar should provide a better estimation, but it is not yet available for the whole European territory, and its temporal resolution may be insufficient to capture the dynamics of these covers.

Concerning the final European fuel map (1 km spatial resolution), only 20 out of the 24 possible first-level fuel types were mapped because the remainder did not cover a continuous large enough area to be represented at 1 km resolution. The herbaceous cropland and the closed needleleaf evergreen forests are the most extended fuel types in Europe, which is related to the land use activities of the European society and the natural distribution of vegetation species due to bioclimatic conditions (García-Martín et al., 2001). Also, the large extension of forest fuel types constitutes an increasing potential risk in light of the growing trends of land abandonment, particularly in remote areas; forests with a high surface fuel load can more easily turn into crown fires (Scott and Reinhardt, 2001; Weise and Wright, 2014), characterised by high intensity, and emit vast amounts of the stored carbon. Urban fuel types are the least represented in Europe, but they are the most dangerous from an economic, societal, and human health point of view (Bowman et al., 2011). Mapping urban fuel types represents an advance of the proposed classification system, as it allows the assessment of residential and non-natural fuels, which can, in turn, help in the identification of anthropic areas where fires can affect human settlements and lives.

Finally, the quantitative assessment of the European fuel map obtained a high overall accuracy of 88.40 %, with average commission errors of 37 % (highest for the nonfuel category and lowest for the forest fuel types) and average omission errors of 29 % (highest for the grassland and lowest for the urban fuel types). Although the obtained accuracy is higher than the one for our main information source, which was the Copernicus GLC map (Tsendbazar et al., 2020), and it surpassed the ideal 85 % minimum overall accuracy, not all fuel types presented the ideal ≥70 % accuracy (Thomlinson et al., 1999). The overall accuracy was higher than the one for the 2019 Copernicus GLC map over Europe (79.9 %), probably due to the validation approach. The confusion matrix is aligned with the confusion matrix of the 2015 global Copernicus GLC maps over Europe (Tsendbazar et al., 2020), when considering most similar categories. The errors in the Copernicus GLC map have been transferred to the European fuel map, as it was our main information source.

With similar accuracies to the 2015 Copernicus GLC map over Europe, forest fuel types present low omission and commission errors, although there is some confusion with shrubland, grassland, and cropland. The shrubland omission and commission errors (mostly confused by the Mediterranean sclerophyllous and xerophilic forest) are significant; however, our validation approach obtained 31 % and 2 % less, respectively, compared to the 2015 Copernicus GLC map. The grassland omission errors (mostly confused by herbaceous cropland) are 15 % higher than the ones for herbaceous vegetation in the 2015 Copernicus GLC map. In addition, grassland commission errors are 17 % lower than in the 2015 Copernicus GLC map. Croplands present higher (+7 % and 11 %) producer and user accuracies than the 2015 Copernicus GLC map, mostly confused with grassland, with the producer accuracy being higher than the user accuracy, as in the Copernicus GLC map. Wet and peat/semi-peat land omission errors are 3 % lower and commission errors are 11 % higher than in the 2015 Copernicus GLC map for herbaceous wetland, in agreement with the observed accuracy tendencies (Tsendbazar et al., 2021). Urban fuel types have the lowest omission error (0 %) and only 25 % commission errors. The nonfuel category errors are mostly referred to pixels over the coastline caused by the different spatial resolutions of the European fuel map and the LUCAS points. This also happens to the rest of the fuel types and is considered to be the main limitation of the validation method. Some validation errors are also caused by the different dates of the input sources and the validation data.

The quantitative assessment of all mapped FirEUrisk fuel types obtained a medium–high overall accuracy of 81.22 %, with average commission errors of 40 % (highest for the high grasslands and tree and shrubland wet and peat/semi-peat land fuel types and lowest for the herbaceous cropland fuel type) and average omission errors of 43 % (highest for the medium grassland fuel type and lowest for the urban continuous fabric fuel type). These results are higher than those of the Copernicus GLC map (Tsendbazar et al., 2020), but neither surpass the ideal 85 % minimum overall accuracy, nor do all fuel types have ≥70 % accuracy (Thomlinson et al., 1999). However, the visual assessment improved the validation method because it considered the entire 1 km2 pixels and not only the area of the LUCAS points. This method could only be applied to a subset of the validation points because of its temporal and human cost compared to the previous validation method. The results are similar to the confusion matrices of the FirEUrisk main fuel types and the Copernicus GLC map over Europe (Tsendbazar et al., 2020), although errors are higher and different due to the dissimilar validation methods and reference data, and confusion appears between fuel types belonging to the same main fuel type. Most errors are due to pixels with a mixed cover of fuel types and low quality of the reference data (unclear and blurred Google images and LUCAS photos and pixels not meeting all ideal conditions for validation – that was needed to have a representative sampling for every fuel type). Input and reference data temporal differences may also have affected the accuracy. The obtained errors present the typical pattern for land cover and vegetation classifications with remote sensing (used to develop the input data), dependent on the separability of the spectral signatures of the land types. This explains why errors are dominant for fuel types belonging to the same main fuel type instead of fuel types from different main fuel types.

Forest fuel types have acceptable accuracy, except for the closed mixed forest, which is highly confused with closed needleleaf evergreen forest. Many errors refer to the omission of open forest, assigned to the closed forest, as in the Copernicus GLC map over Europe (Tsendbazar et al., 2020). Shrubland and grassland fuel type errors are significant, mostly between fuel bed depth categories. Therefore, care must be taken with these results, as estimating fuel bed depth from photos is challenging, and fuel bed depth varies with time. These limitations especially affect grassland due to its low depth and rapid growth and the fact that high grassland is frequently cut. Thus, grassland fuel bed depth is very changeable, so we assume the European fuel map may only be accurate for some periods of the year. We validated the proposed fuel map considering the mean potential fuel bed depth. Moreover, short grassland is generally confused with the herbaceous cropland of fodder crops of agriculturally improved grasslands and temporary pasture such as legumes. Cropland fuel types are the most accurate, with no significant errors. Wet and peat/semi-peat land fuel types have moderate accuracy. The most important confusions are between the tree wet and peat/semi-peat land with other wet and peat/semi-peat land fuel types and between the shrubland wet and peat/semi-peat land with shrubland. The urban continuous fuel type has no omission errors, while some commission errors are in favour of the urban discontinuous fuel type in the outskirts of the residential areas of cities. The urban discontinuous fuel type presents a higher omission than commission errors, which is mostly omitted by cropland in agricultural rural areas. Similar to the confusion matrix for the main fuel types, both commission and omission errors for the nonfuel category are low and relate to mixed pixels.

The different levels of disaggregation of the proposed classification system, and the main fire behaviour characteristics of the diverse fuels, made the crosswalk challenging and did not allow us to assign a specific standard model to each FirEUrisk fuel type. Moreover, the FBFM standard fuel models (Scott and Burgan, 2005) were originally developed for the United States, so care must be taken when using the crosswalk in Europe (Santoni et al., 2011; Salis et al., 2016). From this point of view, our proposed approach can be improved in specific areas if customised information and data on given fuel types are available (Arca et al., 2007; Fernandes, 2009; Duguy Pedra et al., 2015; Kucuk et al., 2015; Ascoli et al., 2020). In other words, we propose a generic crosswalk scheme, but users are free to wisely choose or modify the best-fitting standard fuel models according to their study area and expertise or to use different parameters from the standard ones if they have better information for given study areas. Moreover, the main limitation of the crosswalk scheme relies on the reference to general bioclimatic regimes, which is not able to fully consider all inherent differences among European regions in terms of fuel characteristics, while moisture values can be spatially modified according to the specific status of each fuel type.

This work represents one of the first attempts to adopt a standardised fuel model mapping approach over Europe, similar to the National Fire Danger Rating fuel model products that have been available since the 1990s for the continental United States (see, for instance, https://www.wfas.net/index.php/nfdrs-fuel-model-static-maps-44, last access: 7 March 2023). Work is in progress to develop higher-resolution products over Europe that combine a set of remote sensing tools and data. This latter development at the European scale is highly complicated by the huge heterogeneity in the availability of high-quality and high-resolution ground and measured data, which vary a lot among and within the different regions.

The FirEUrisk fuel classification system can provide a number of insights and information for wildfire risk monitoring and assessment at the European scale, including fuel parameters, such as dead and live surface fuel load, the surface to area volume ratio, or surface fuel bed depth. This is mostly related to the identified fuel categories crosswalk to the FBFM system (Scott and Burgan, 2005), which is specifically designed for the above purpose. In fact, the parameters included in each FBFM model allow the characterisation of surface fuels and can serve as a baseline for surface wildfire spread and behaviour modelling. The full surface fuel set information needed to run fire propagation models can be extracted from the crosswalk to the FBFM and complemented with other canopy fuel parameters (such as crown base height or crown bulk density) and other necessary input data (e.g. weather conditions, topography, and ignitions) to run fire spread models (e.g. FlamMap, Finney, 2006; FARSITE, Finney, 2004), as embedded in FlamMap 6.2 (https://www.firelab.org/project/flammap, last access: 7 March 2023). This should be the subject of an extension of this paper and could be based on the calibration of models that estimate canopy fuel parameters using airborne and satellite lidar systems for which regional airborne lidar would be key to consider the heterogeneity of European fuels before using the global satellite lidar data for the continental scale.

The fuel map is also expected to provide estimations of fire-caused carbon emissions and pollution and estimations of biomass consumption. The Consume model (Prichard et al., 2006) could be used for this if a crosswalk to FCCS fuels is previously made, including the necessary fuel parameters such as the combustion percentage. In addition, the FirEUrisk fuel map would be useful for regions that do not have fuel cartography. The mapped fuel types and the fuel parameters obtained from the crosswalk to FBFM can serve as input for fire propagation models and help rate fire danger and risk conditions. It is also important to note that the maps of fuel parameters at the European scale are examples of what can be done, but the crosswalk is intended to be useful for areas where technologies and resources such as lidar data are not available.

Overall, we highlight that the main use of the map is providing a dataset able to rate fire danger and risk conditions across large geographic areas, while the application of wildfire spread models to very local scales or small areas may pose limitations in the quality of outputs due to the low resolution (1 km resolution) of the fuel input layer.

Finally, although it has been developed for European conditions, our methodology has the potential to be applied to other regions. The proposed fuel classification system could be used in several fire applications and adapted anywhere in the world to further extend the fuel subcategories wherever required. The classification of fuel types is dependent on existing land cover and biogeographic data, but it can also be directly estimated from satellite data, either at a coarse resolution for continental areas or at a higher resolution for smaller territories. The fuel parameterisation can also be based on other standard fuel models, such as the NFFL or the FCCS, but it can also rely on ground measurements or more detailed regional fuel characteristics. In any case, it is important to emphasise the need for estimating fuel parameters in order to use the fuel type products for quantitative estimations of fire risk, behaviour, and effects. This is a key aspect of the FirEUrisk project and a crucial point towards wildland fire prevention across the European Union.

The resulting European fuel map (circa 2019; 1 km spatial resolution) in one single-band categorical raster layer in a GeoTIFF format is publicly available at 10.21950/YABYCN (Aragoneses et al., 2022a), and a product user manual (PUM; Aragoneses et al., 2022b) is available from e-cienciaDatos (https://edatos.consorciomadrono.es/dataset.xhtml?persistentId=doi:10.21950/YABYCN, last access: 7 March 2023).

This paper, developed in the framework of the European FirEUrisk project, presents a new hierarchical fuel classification system for surface and canopy fuels adapted to the European conditions, in addition to methods to map those categories and assign them fuel parameters. The final European fuel map contains 20 fuel types, including both surface and canopy fuel types. The estimated overall accuracy was 88 % for the main fuel types and 81 % for all mapped fuel types. Finally, the paper shows an example of a crosswalk between the proposed fuel types and standard fuel models, in this case the Fire Behavior Fuel Models (FBFMs; Scott and Burgan, 2005), that provides a full set of surface fuel parameters useful for surface fire behaviour modelling. Our approach, based on expert knowledge, GIS, existing land cover datasets, biogeographic data, and bioclimatic modelling, could be readily applied to other regions.

The results of this study constitute the first step towards a risk-wise landscape and fuel mapping development across Europe, which will help in integrated, strategic, coherent, and comprehensive decision-making for fire risk prevention, assessment, and evaluation. The results have wide applicability because they meet the actual unfulfilled fuel mapping needs in Europe, including (1) the development of a fuel classification system specifically designed for European conditions, which allows us to not rely on external classifications that should be only applied to the regions for which they were developed, (2) enabling coordination, by integrating fuel mapping at different spatial scales and across European regions through a common fuel legend with hierarchical levels, (3) multipurpose use, such as prevention, propagation, behaviour, emissions, and suppression, (4) mapping fuel types not previously considered at European scale that are key for protecting people and the environment from the devastating effects of fires such as forest canopy fuels (key for crown and extreme fires), wet and peat/semi-peat land fuels (key for emissions), and urban fuels in the wildland–urban interface (key for human and socioeconomic safety), (5) the generation of an updated European-specific fuel map, compared to the EFFIS fuel map from year 2000 (European Forest Fire Information System, 2017), and (6) the preliminary surface fuel parameterisation for Europe that can be used for estimating fuel parameters whenever there is no suitable input data available. Overall, the existence of updated land cover datasets and bioclimatic models for the European territory is limiting, and work is still needed to parameterise canopy fuels. The results of this work are part of the new FirEUrisk integrated three-part perspective of fire risk, whose strategy is meant to shift the thinking of wildfire management by looking simultaneously at fire assessment, reduction, and adaptation from a common scheme.