Remote Sensing of Environment Manuscript Draft. Manuscript Number:

Transcription

1 Remote Sensing of Environment Manuscript Draft Manuscript Number: Title: Fine resolution fuel characterization for the fragmented ecosystems of Pollino National park in the South of Italy by using Prometheus model and satellite ASTER data Article Type: Full length article Section/Category: Keywords: ASTER data, fuel types, fire Manuscript Region of Origin: Abstract:

2 Abstract Abstract. Fuel types is one of the most important factors that should be taken into consideration for computing spatial fire hazard and risk and simulating fire growth and intensity across a landscape. In the present study, forest fuel mapping is considered from a remote sensing perspective. The purpose is to delineate forest types by exploring the use of satellite remote sensing ASTER (Advanced Spaceborne Thermal Emission and Reflection Radiometer) imagery. In order to ascertain how well ASTER data can provide an exhaustive classification of fuel properties a sample area characterized by mixed vegetation covers and complex topography was analysed. The study area has an extension at around 60 km2 and is located inside Pollino National Park in the South of Italy. Fieldwork fuel type recognitions, performed before, after and during the acquisition of remote sensing ASTER data, were used as ground-truth dataset to assess the obtained results. The method comprised the following three steps: (I) adaptation of Prometheus fuel types for obtaining a standardization system useful for remotely sensed classification of fuel types and properties in the considered Mediterranean ecosystems; (II) model construction for the spectral characterization and mapping of fuel types based on a maximum likelihood (ML) classification algorithm; (III) accuracy assessment for the performance evaluation based on the comparison of ASTER-based results with ground-truth. Results from our analysis showed that the use of remotely sensed ASTER data provided a valuable characterization and mapping of fuel types being that the achieved classification accuracy was higher than 80%.

3 * Manuscript Fine resolution fuel characterization for the fragmented ecosystems of Pollino National park in the South of Italy by using Prometheus model and satellite ASTER data Rosa Lasaponara 1, Antonio Lanorte 1 1 National Research Council, Institute of Methodologies of Environmental Analysis, C. da S. Loja, Tito Scalo, Potenza 85050, Italy. Abstract. Fuel types is one of the most important factors that should be taken into consideration for computing spatial fire hazard and risk and simulating fire growth and intensity across a landscape. In the present study, forest fuel mapping is considered from a remote sensing perspective. The purpose is to delineate forest types by exploring the use of satellite remote sensing ASTER (Advanced Spaceborne Thermal Emission and Reflection Radiometer) imagery. In order to ascertain how well ASTER data can provide an exhaustive classification of fuel properties a sample area characterized by mixed vegetation covers and complex topography was analysed. The study area has an extension at around 60 km2 and is located inside Pollino National Park in the South of Italy. Fieldwork fuel type recognitions, performed before, after and during the acquisition of remote sensing ASTER data, were used as ground-truth dataset to assess the obtained results. The method comprised the following three steps: (I) adaptation of Prometheus fuel types for obtaining a standardization system useful for remotely sensed classification of fuel types and properties in the considered Mediterranean ecosystems; (II) model 1

4 construction for the spectral characterization and mapping of fuel types based on a maximum likelihood (ML) classification algorithm; (III) accuracy assessment for the performance evaluation based on the comparison of ASTER-based results with groundtruth. Results from our analysis showed that the use of remotely sensed ASTER data provided a valuable characterization and mapping of fuel types being that the achieved classification accuracy was higher than 80%. 1. Introduction Wildland fires are considered one of the most important disturbance factors in the natural ecosystems of the Mediterranean regions, where every year, around 45,000 forest fires break out causing the destruction of about 2,6 million hectares (FAO, 2001). Several studies (see, for example, Vila et al. 2001) dealing with the effects of fires on the vegetation within the Mediterranean basin found that fires induce significant alterations in short as well as long-term vegetation dynamics. Prevention measures, together with early warning and fast extinction, are the only methods available that can support fire fighting and limit damages caused by fires especially in regions with high ecological value, dense populations, etc. In order to limit fire damage, fire agencies need to have effective decision support tools that are able to provide timely information for quantifying fire risk. In particular, fire managers need information concerning the distribution, amount and condition of fuels in order to improve fire prevention and modelling fire spreading and intensity. Geographic Information Systems (GIS) and Remote Sensing (RS) are considered useful tools for supporting prevention activities (see, for example, Chuvieco et al. 2004). Remote sensing can provide valuable data on type (namely distribution and amount of fuels) and status of 2

5 vegetation in a consistent way at different spatial and temporal scales. Since the description of fuel proprieties is usually very complex, fire managers have tried to summarize the physical parameters and spatial distribution of fuel in different classes also known as fuel models (Anderson 1982, Burgan and Rothermal 1984). More specifically, a fuel model has been defined as "an identifiable association of fuel elements of distinctive species, form, size, arrangement, and continuity that will exhibit characteristic fire behaviour under defined burning conditions" (Merrill and Alexander 1987). The spatial distribution of the fuel characteristics can be displayed as fuel type maps. Northern Forest Fire Laboratory (NFFL) system (Albini 1976) is the most common and well-know fuel model that was developed taking the vegetation structure and characteristic of the North-American floras in to account. Recently, in the framework of the Prometheus project (1999), a new fuel type classification system was specifically developed to better represent the fuel characteristic of the Mediterranean ecosystems ( Algosystems SA, Greece). This classification is principally based on the height and density of fuel, which directly influence the intensity and propagation of wildfire. Due to the complex nature of fuel characteristic a fuel map is considered one of the most difficult thematic layers to build up (Keane et al 2000, Keane et al 2000) especially for large areas. Aerial photos have been the most common remote sensing data source traditionally used (Morris 1970, Muraro 1970, Oswald et al. 1999) for mapping fuel types distribution. Nevertheless, remote sensing multispectral data can be an effective data source available at different temporal and spatial scales that can be fruitfully adopted for 3

6 building up fuel type maps from global, region down to local scale. For this purpose, up to now, several satellite sensors have been used in last decades. For example, NOAA- AVHRR (Advanced Very High Resolution Radiometer) data were used by McKinley et al.(1985) for mapping fuel types in Western United States. Landsat Thematic Mapper data were used for mapping fuels models in Yosemite national Park, USA (Van Wagtendonk and Root 2003); and in Spain (Cohen 1989 in California; Riaño et al Salas and Chuvieco 1994). A multisensor approach based on Spot and Landsat imager was adopted by Castro and Chuvieco (1998) to perform a classification of fuel types for Chile by using an adapted version of Anderson s system. The accuracies obtained from these researches ranged from 65% to 80% (Chuvieco 1999). The accuracy level is strongly related with fuel presence and spatial distribution (how many and where) and with specific environmental conditions (topography, land cover heterogeneity, etc.).the importance of using multisensor data source to map fuel model was emphasised by many authors (see, for example, Keane et al. 2001). Although the recognized feasibility of satellite sensors traditionally used for the remote characterization of fuel types, the advent of new sensors with improved spatial and spectral resolutions may improve the accuracy (Chuviceo and Congalton, 1989) and reduce the cost (Zhu and Blumberg, 2001) of forest fire fuel mapping. Up to now, fire researchers did not paid enough attention to the potentiality of using remote sensing ASTER (Advanced Spaceborne Thermal Emission and Reflection Radiometer) data to map fuel types and properties. This research aims to investigate the usefulness of ASTER data to characterize and map fuel types in fragmented ecosystems, such as those found in the Mediterranean 4

7 basin. This objective is achieved by using Prometheus model coupled with ASTER data that were analysed by using Maximum Likelihood (ML) classifier for a test case (located in the south of Italy) that is highly representative of Mediterranean like ecosystems. 2. Study area The selected study area (figure 1 a and b) extends over a territory of about hectares inside Pollino National Park in the Basilicata Region (Southern Italy). Figure 1a shows the location of Pollino Park and Figure 1 b shows a red green blu (RGB) composition of ASTER spectral channels performed for the study area. The study area constitutes a complex morphological unit in which mountain landscapes are close to Sinni river alluvial plan. It is characterized by complex topography with altitude varying from 400 m to 1900 m above sea level (asl) and mixed vegetation covers. Between 400 and 600 m natural vegetation is constituted by the Mediterranean scrubs, xeric prairies and Mediterranean shrubby formations. In the strip included between 600 and m the characteristic vegetation is represented by poor populations of Quercus pubescens and from extensive woods of Turkey oaks (Quercus cerris); evident degradation forms are present, in the form of xerophytic prairies and substitution bushes. The higher horizons are constituted by beech woods (Fagus sylvatica) which arrive up to 1900 m: the deforested areas in this strip are generally engaged by mesophytic prairies used for pasture. 5

8 3. Data Analysis 3.1 Dataset ASTER ASTER is a high resolution imaging instrument that is flying on the Earth Observing System (EOS) Terra satellite. It has the highest spatial resolution of all five sensors on Terra and collects data in the visible/near infrared (VNIR), short wave infrared (SWIR), and thermal infrared bands (TIR). Each subsystem is pointable in the crosstrack direction. The VNIR subsystem of ASTER is quite unique. One telescope of the VNIR system is nadir looking and two are backward looking, allowing for the construction of 3- dimensional digital elevation models (DEM) due to the stereo capability of the different look angles. ASTER has a revisit period of 16 days, to any one location on the globe, with a revisit time at the equator of every four days. ASTER collects approximately eight minutes of data per orbit (rather than continuously). Among the 14 ASTER bands we only considered the 3 channels in the VNIR region and 6 channels in the SWIR region, while the TIR channels were excluded. The ASTER data used for this study were sensed on July, Additionally, photos and air photos were obtained for the investigated area immediately before and after the acquisition of satellite ASTER data. Fieldwork fuel typing were performed using a global position system (GPS) for collecting geo-position data (latitude and longitude). Air photos and fieldwork fuel types were used as a groundtruth dataset firstly to identify the fuel types defined in the context of Prometheus system, and secondly, to evaluate performance and results obtained for the considered test area from the ASTER data processing. 6

9 3.2 Prometheus adaptation The seven fuel type classes standardized in the context of Prometheus system (see Table 1) were detailed identified and carefully verified for the study area on the basis of field works performed before, during and after the acquisition of ASTER remote sensing data. In particular, photos and air photos, taken for the investigated region immediately before and after the acquisition of ASTER data, along with the fuel types recognized in the field were used for this purpose. Figure 2 and Table 2 show the results obtained from the adaptation of Prometheus system to the characteristics and properties of fuel types present in the investigated test area. Significant patches corresponding to areas representative for each fuel class were carefully identified over the ASTER images by using geo-position data (latitude and longitude) collected during the ground surveys by means of a GPS positioning system. Pixels relating to these areas were exploited for performing the selection of adequate Region of Interest point (Ground-Truth dataset) for the seven classes (fuel types) with the addition of one class concerning areas having no Fuel and 1 class concerning areas having unclassified pixels. The sample points of Ground-Truth dataset were selected in the same areas subject to direct check on field in order to be used firstly to identify the fuel types defined in the context of Prometheus system, and secondly, to evaluate performance and results obtained for the considered test area from the ASTER data processing. For this reason, pixels corresponding to the given Ground-Truth areas were subdivided in to testing data and training data through randomization of the pixels to 50% for every class. 7

10 3.3 Model construction and comparison The mapping of fuel types was obtained by using a supervised classification based on Maximum Likelihood (ML) algorithm. The ML classifier is considered one of the most important and well-known image classification methods due to its robustness and simplicity. It is wide used in vegetation and land cover mapping. Moreover, it was also tested for fuel model distribution (Riano and Chuvieco, 2002). The ML method quantitatively evaluates the variance and covariance of the spectral signatures when classifying an unknown pixel assuming at the same time a Gaussian distribution of points forming a cluster of a vegetation class. Under this assumption the distribution of a class is described by the mean vector and covariance matrix which is used to compute the statistical probability of a given pixel value being a member of a particular class. The probability for each class is calculated and the class with the highest probability is assigned the pixel (Lillesand and Kiefer 2000). As above reported, the ML classifier is based on the assumption that different variables used in the computation are normally distributed. This assumption is generally considered acceptable for common spectral response distribution. In our case, on the basis of ground surveys and air photos, we selected the Region Of Interest (ROI) corresponding to the considered seven fuel types, plus two additional classes related to no fuel and unclassified regions. Pixels belonging to each of the considered ROI were randomly separated into training data and testing data, used for the ML and accuracy evaluation respectively. 8

11 4 RESULTS Figure 3 shows the mapping of fuel types obtained for the investigated test area from the ASTER images. Such map presents very high user's accuracy. For the accuracy assessment we consider the producer accuracy, user accuracy, and overall accuracy, that are defined as follows. The producer s accuracy is a measure indicating the probability that the classifier has correctly labelled an image pixel, for example, into Fuel Type 1 class given that, on the basis of ground recognition such a pixel belongs to Fuel Type 1 class. The user s accuracy is a measure indicating the probability that a pixel belongs to a given class and the classifier has labelled the pixel correctly into the same given class. The overall accuracy is calculated by summing the number of pixels classified correctly and dividing by the total number of pixels. Finally, the kappa statistics (K) was also considered. It measures the increase in classification accuracy over that of pure chance by accounting for omission and commission error (Congalton and Green, 1998). Overall accuracy is computed as the sum of the number of observations correctly classified (class1, as class 1, class 2 as class 2, etc.) divided by the total number of observations. This is equivalent to the diagonal of a square contingency table matrix divided by the total number of observations described in that contingency table (Congalton and Green, 1998). Table 3 shows the accuracy coefficients as well as the omission and commission errors. The achieved overall accuracy was 82,22%. Thus, showing that the use of remotely sensed ASTER data at high spatial and spectral resolution provided a valuable characterization and mapping of fuel types. 9

12 Fuel type 2 (47.00%) and fuel type 3 (58.80%) show the worst producer accuracy; whereas fuel type 2 (44.90%) and fuel type 4 (54.32%) show the worst user accuracy. This is mainly due to due mixing of different fuel types of concern. As regards Fuel type 7 and Fuel type 4 the mixing between the two classes consists in a remarkable pixels "exchange because ML Classifier "moves" 14,57% of the pixels attributed in the Ground-Truth Dataset to Fuel type 7 toward Fuel type 4 and 11,29% to Fuel type 4 toward Fuel type 7. The mistake goes above all attributed what to the difficulty to select distinct Ground-Truth Points for the two classes, which present a few analogous vegetation elements, being different on the structural plan. The major classifications problems come from the high mixing between Fuel Type 3 with Fuel Type 4 and Fuel Type 2: in particular occur a remarkable pixels transition (almost 19% of the pixels attributed in the Ground-Truth Dataset to Fuel type 3) from Fuel Type 3 towards Fuel type 4 while the pixels "transfer" in adverse sense is much more limited (less than 6%). Instead, between Fuel type 3 and Fuel type 2 there is an nearly analogous exchange of a pixels number (around 11%). Also in this case is necessary to consider that Fuel type 3 presents a few analogous vegetation elements as with Fuel Type 2 as with Fuel Type 4, being different on the structural plan: therefore the mixing turns out unavoidable. Besides, the level of classification accuracy is substantially determined by the mixing between Fuel type 2 with Fuel type 1 and Fuel type 6: ML Classifier "moves" 16% of the pixels attributed in the Ground-Truth Dataset to Fuel type 2 toward Fuel type 1 and less than 12% to Fuel type 2 toward Fuel type 6. In this case, the mistake must be attributed to 10

13 the fact that this fuel type has a subpixel distribution, thus making impossible the selection of adequate ROI at the pixel level. Finally, Aster ML shows very high accuracy levels in consideration of the high variability of vegetation typologies. 5. Conclusions Multispectral ASTER data acquired in July, 2004 were analysed for a test area of southern Italy to ascertain how well remote sensing data can characterize fuel type and map fuel properties. Fieldwork fuel type recognitions, performed at the same time as remote sensing data acquisitions, were used as ground-truth dataset to assess the results obtained for the considered test area. Results from our analysis showed that the use of remotely sensed data at high spatial and spectral resolution provided a valuable characterization of fuel types being that the classification accuracy was higher than 82%. Results obtained from these investigations can be directly extended to Mediterranean like ecosystems. The approach proposed in this work can be fruitfully applied to different remote sensed data, such as Quickbird, Ikonos, Landsat Thematic Mapper, or Enhanced Thematic Mapper, NOAA- AVHRR, TERRA-AQUA MODIS, characterized by different spatial and spectral resolution for mapping fuel properties at different spatial scale from landscape to regional level as it is highly requested by National and International Environmental protection Agency, Forestry and Environmental managers. 11

14 References Adams, J. B., Smith, M. O., and Johnson, P. E., (1985), Spectral mixture modelling: a new analysis of rock and soil types at the Viking Lander I site. Journal of Geophysical Research, 91, Albini, F.A., (1976), Estimating wildfire behaviour and effects. General Technical Report. USDA Forest Service, Intermountain Forest and Range Experiment Station INT-30. Ogden, UT, pp Anderson, H.E., (1982), Aids to determining fuels models for estimating fire behaviour. USDA Forest Service, Intermountain Forest and Range Experiment Station General Technical Report INT-122. Ogden, UT. 22 pp. Boardman, J.W., (1998), Leveraging the high dimensionality of AVIRIS data for improved sub-pixel target unmixing and rejection of false positives: mixture tuned matched filtering. Summaries of the Seventh JPL Airborne Geoscience Workshop. JPL Publication 97-1, NASA Jet Propulsion Lab., Pasadena, Calif. Boardman, J.W., Kruse, F.A., and Green, R. O., (1995), Mapping target signatures via partial unmixing of AVIRIS data. Summaries of the Fifth JPL Airborne Geoscience Workshop, JPL Publication 95-1, NASA Jet Propulsion Lab., Pasadena, Calif. Burgan, R., and Rothermal, R.C., (1984), BEHAVE: fire behaviour prediction and fuel modelling system-fuel subsystem. USDA Forest Service General Technical Report INT pp. 12

15 Castro, R. and Chuvieco E., (1998), Modelling Forest Fire danger From Geographic Information System. Geocarto International, 13, Chuvieco, E., (1999), Remote sensing of large wildfires in European Mediterranean basin. Springer-Verlag, Berlin Chuvieco, E., Cocero, D., Aguado, I., Palacios, A., and Preado, E., (2004), improving burning efficiency estimates through satellite assessment of fuel moisture content. Journal of Geophysical Research, 109, doi: /2003jd Cohen, W. B., (1989), Potential utility of the TM tasseled cap multispectral data transformation for crown fire hazard assessment. ASPRS/ACSM annual convention proceedings: Agenda for the 90 s. Volume 3. Baltimore Maryland. pp FAO, (2001). Global forest fire assessment Forest Resources Assessment Programme, working paper n.55, Harsanyi, J.C., and Chang, C., (1994), Multispectralimage classification and dimensionality reduction: an orthogonal subspace projection approach. IEEE Trans. Geosci. Remote Sens., 32, Keane, R.E., Mincemoyer, S.A., Schmidt, K.A., Long, D.G., and Garner, J.L., (2000), Mapping vegetation and fuel for fire management on the Gila National Forest Complex, New Mexico. USDA Forest Service General Technical Report RMRS- GTR-46-CD. 13

16 Keane, R.E, Burgan, R., and Van Wagtendonk, J., (2001), Mapping wildland fuels for fire management across multiple scales: Integrating remote sensing, GIS, and biophysical modelling. International Journal of Wildland Fire, 10 (3-4), Ichku, C., and Karnieli, A., (1996), A review of mixture modelling techniques for subpixel land cover estimation. Remote Sensing Reviews, 13, Lillesand, T. M., and Kiefer, R.W., (2000), Remote sensing and image interpretation (New York: Wiley). Mckinley, R. A., Chine, E. P. and Werth, L. F., (1985), Operational fire fuels mapping with NOAA-AVHRR data. American Society for Photogrammetry and Remote Sensing, Falls Church, Virginia. pp Merrill, D. F., and Alexander, M. E., (1987), Glossary of forest fire management terms. National Research Council of Canada, Committee for Forest Fire Management, Ottawa. Morris, W. G.,(1970), Photo inventory of fine logging slash. Photogrammetric Engineering 36, Muraro, S. J., (1970), Slash fuel inventories from 70 mm low-level photography. Ottawa, Ontario, Canadian Forest Service: 63. Oswald, B. P., Fancher, J. T., Kulhavy, D. L., and Reeves, H. C.,(1999), Classifying fuels with Aerial Photography in East Texas. International Jurnal of Wildland Fire 9 (2), Parker Williams, A., and Hunt JR., E.R., (2002), Estimation of leafy spurge cover from multispectralimagery using mixture tuned matched filtering. Remote Sensing of Environment, 82,

17 Riaño, D., and Chuvieco, E., (2002), Generation of fuel type maps from Landsat-TM images and auxiliary data in Mediterranean ecosystem. Alcalá de Henares (Spain), Department of Geography, Alcalá de Henares University. Salas, J., and Chuvieco, E., (1994), Geographic information system for wildland fire risk mapping. Wildfire 3(2), 7-13 Settle, J. J., and. Drake, N. A., (1993), Linear mixing and the estimation of round over proportions. Int. J. Remote Sensing, 14(6), Smith, M. O., Ustin, S. L., Adams, J. B., & Gillespie, A.R., (1990),Vegetation in deserts: I. A regional measure of abundance from multispectral images. Remote Sensing of Environment, 31, Stow, D., Hope, A., Mckinsey, D., and Pray, H., (1993), Deriving dynamic information on fire fuel distributions in southern California chaparral from remote sensed data. Landscape and Urban Planning 24, UNCCD, (1994) United Nations Convention to Combat Desertification, report, Paris,. Van Der Meer, F., (1995), Spectral unmixing of Landsat Thematic Mapper data. Int. J. Remote Sensing, 16(16), Van Der Meer, F., (1999), Spatial Statistics for Remote Sensing, chapter Image Classification through Spectral Unmixing, pp Kluwer Academic Publishers, first edition. Van Wagtendonk, J. W., and Root, R.R., (2003), The USE of multitemporal Landsat Normalized Difference Vegetation Index (NDVI) data for mapping fuels models in Yosemite national Par k, USA, International Journal of remote Sensing, 24,

18 Vila M., Lloret, F., Ogheri, E., and Terradas, J., (2001), Positive fire-grass feedback in Mediterranean. basin shrubland. Forest Ecology and Management, 147, Yool, S. R., Eckhardt, D. W., Estes, J. E., and Cosentino, M. J., (1985), Describing the brushfire hazard in the southern California. Ann. Assoc. American Geogr. 75,

19 Figure and Table Caption Figure 1a Figure 1b Figure 2 Figure 3 Table 1 Table 2 Table 3 Location of the study area RGB composition of ASTER spectral channels for the study area. Prometheus fuel model Fuel map obtained from the processing of ASTER data Fuel type classification developed (Prometheus Project 1999) for Mediterranean ecosystems in framework of the Prometheus project Fuel type and vegetation typologies Accuracy coefficients 17

20 Tables Table 1 Fuel type classification developed for Mediterranean ecosystems in the framework of Prometheus project (Prometheus Project 1999) Fuel Type class Fuel Type description in terms of percentage of cover 1 Ground fuels (cover >50%) grass 2 Surface fuels (shrub cover >60%, tree cover <50%) 3 Medium-height shrubs (shrub cover >60%, tree cover <50%) 4 Tall shrubs (shrub cover >60%, tree cover <50%) 5 Tree stands (>4 m) with a clean ground surface (shrub cover <30%) 6 Tree stands (>4 m) with medium surface fuels (shrub cover >30%) Fuel Type description in terms of vegetation typology grassland, shrub land (smaller than m and with a high percentage of grassland), and clear cuts, where slash was not removed shrubs between 0.6 and 2.0 m high shrubs (between 2.0 and 4.0 m) and young trees resulting from natural regeneration or forestation the ground fuel was removed either by prescribed burning or by mechanical means. This situation may also occur in closed canopies in which the lack of sunlight inhibits the growth of surface vegetation the base of the canopies is well above the surface fuel layer (>0.5 m). The fuel consists essentially of small shrubs, grass, litter, and duff (the layer of decomposing organic materials lying immediately above the mineral soil but below the litter layer of freshly fallen twigs, needles, and leaves; the fermentation layer). 7 Tree stands (>4 m) with heavy surface fuels (shrub cover >30%) stands with a very dense surface fuel layer and with a very small vertical gap to the canopy base (<0.5 m)

21 Table 2. Fuel type and vegetation typologies Fuel types Fuel type Fuel type and vegetation typologies Roads and houses no fuel Plowed and bare soils Woody cultivations Calcareous cliffs and detritus Areas at present or permanently presenting soils with no vegetation cover because of plowing or erosion phenomena Arboreal cultivations (mainly orchards, olive-groves and vineyards) Emergine rocks, landslides and generally melted sediments fuel type 1 Course and water bodies Xerophytic Prairies Mesophytic Prairies Natural water and artificial courses besides natural water stretches and artificial basins Grassy vegetation of primary and secondary origin with a prevalence of Bromus sp. Grassy vegetation developing when water availability is good. Middle altitude prairies and peak prairies a prevalence of Cynosurus cristatus fuel type 2 fuel type 3 fuel type 4 Substitution bushes Uncultivated soils, ferns, field boundary vegetation, woods, meadows, roads Small shrubs with predominant presence of Cistus sp. Vegetation growing in areas no longer agriculturally used or in any case at borders of lands used for a definite and consolidate purpose Substitution bushes, broom vegetation Mediterranean formations of secondary origin with predominant presence of Prunus sp., Spartium junceum transition states towards forestry vegetation referable to mixed oak woods Mediterranean scrubs, garigues and shrubby prairies Vegetation types referable to evergreen sclerophylls mainly degradated, mediterranean formations with prevalence of Medium-height shrubs and secondary formations scarcely covering areas no longer used for pasture Mediterranean scrubs Vegetation types referable to evergreen sclerophylls mediterranean formations with prevalence of tall shrubs Beech woods and Mixed oak woods Young trees or small height formations of Beech woods and Mixed oak woods fuel type 5 fuel type 6 Ilex woods Beech woods Reforestation conifer woods Oak woods Chestnut woods Wood formations with a prevalent presence of Quercus ilex, of wich there are in the study area scarse pure stands at altitude lower than 600 m Forestry formations with predominant presence of Fagus sylvatica of great relevance in the study area since these trees cover most part of the belt between m. Areas wich have been reforested mainly with Pinus nigra and Pinus halepensis Forestry formations in which, according to the altitude, with prevalence of Quercus pubescens in lower areas and Quercus cerris and Quercus frainetto in higher areas up to Chestnut tree formations (Castanea sativa) fuel type 7 Mixed oak woods and reforesting trees Forestry formations in which Quercus sp. is associated with numerous other forestry species belonging to Acer, Fraxinus, Ostrya, Alnus, Sorbus, Malus, Crataegus, Pinus, Picea, Abies.

22 Table 3. Confusion Matrix Class Producer Accuracy(%) User Accuracy.(%) Fuel type Fuel type Fuel type Fuel type Fuel type Fuel type Fuel type No fuel Unclassified ASTER Overall Accuracy = 82.22% K coefficient=0.78

23 Figure 1 Figure 1a. Arrow and black box indicate the study area Figure 1b RGB composition of ASTER spectral channels for the study area.

24 Figure 2 > 60 % Grass Fuel type 1 Average height ( m) >60% Shrubs & < 50% Trees >4m Fuel type 2 Average height ( m) Fuel type 3 Average height ( m) Fuel type 4 >50% Trees >4m < 30% Shrubs Fuel type 5 >50% Trees >4m > 30% Shrubs Average height difference between shrubs and trees > 0.50m Fuel type 6 Average height difference between shrubs and trees < 0.50 m Fuel type 7

25

26 Figure 3 Unclassified No Fuel Fuel Type 1 Fuel Type 2 Fuel Type 3 Fuel Type 4 Fuel Type 5 Fuel Type 6 Fuel Type 7 Figure 3 fuel map obtained from the processing of ASTER data