[EN] In recent decades, the feasibility of using terrestrial laser scanning (TLS) in forest inventories was investigated as a replacement for time-consuming traditional field measurements. However, the optimal acquisition of point clouds requires the definition of the minimum point density, as well as the sensor positions within the plot. This paper analyzes the effect of (i) the number and distribution of scans, and (ii) the point density on the estimation of seven forest parameters: above-ground biomass, basal area, canopy base height, dominant height, stocking density, quadratic mean diameter, and stand density index. For this purpose, 31 combinations of TLS scan positions, from a single scan in the center of the plot to nine scans, were analyzed in 28 circular plots in a Mediterranean forest. Afterwards, multiple linear regression models using height metrics extracted from the TLS point clouds were generated for each combination. In order to study the influence of terrain slope on the estimation of forest parameters, the analysis was performed by using all the plots and by creating two categories of plots according to their terrain slope (slight or steep). Results indicate that the use of multiple scans improves the estimation of forest parameters compared to using a single one, although using more than three to five scans does not necessarily improves the accuracy. Moreover, it is also shown that lower accuracies are obtained in plots with steep slope. In addition, it was observed that each forest parameter has a strategic distribution depending on the field of view of the TLS. Regarding the point density analysis, the use of 1% to 0.1% (¿136 points·m¿2) of the initial point cloud density (¿37,240.86 points·m¿2) generates an R2adj difference of less than 0.01. These findings are useful for planning more efficient forest inventories, reducing acquisition and processing time as well as costs., This research has been funded by the project PID2020-117808RB-C21 MCIN/AEI/10.13039/501100011033 and by the grant PEJ2018-002924-A Fondo de Garantia Juvenil en I+D+i ESF Investing in your future.

[ES] Los bosques son uno de los ecosistemas más críticos de la Tierra, cubriendo aproximadamente el 31% de la superficie terrestre y desempeñando un papel vital como sumideros de carbono, reservas de biodiversidad y reguladores del clima. Almacenan alrededor del 45% del carbono terrestre y albergan la mayor parte de la biodiversidad del planeta, lo que hace que su conservación sea esencial para el equilibrio ecológico. Sin embargo, estos ecosistemas están cada vez más amenazados por los incendios forestales, que cumplen un doble papel en la dinámica de los bosques: actúan tanto como reguladores naturales como potenciales destructores. La modelización avanzada del fuego es una herramienta clave para mejorar la predicción y mitigación de incendios. Los modelos de incendios 3D, basados en modelos de dinámica de fluidos computacional (CFD, por sus siglas en inglés), requieren datos de entrada de alta resolución sobre la carga de combustible, la estructura de la vegetación y la composición de especies. En este sentido, uno de los mayores desafíos en la modelización de incendios es la obtención de datos precisos y de alta resolución sobre las características de la vegetación a escala local. Las tecnologías de teledetección, en particular la fotogrametría aérea digital con UAV (UAV-DAP, por sus siglas en inglés), ofrecen una solución rentable y de alta resolución para caracterizar la estructura forestal. Mediante UAV-DAP, es posible obtener nubes de puntos que permiten la segmentación individual de plantas, la clasificación de especies y la extracción de características geométricas y espectrales clave para la dinámica del fuego. Estos datos pueden utilizarse para estimar variables relacionadas con los incendios. Sin embargo, a pesar de su potencial, UAV-DAP sigue estando infrautilizado debido a sus limitaciones técnicas inherentes y a la falta de metodologías y software especializados que permitan aprovechar plenamente tanto su información geométrica tridimensional como su contenido espectral. Esta tesis tiene como objetivo mejorar las técnicas de procesamiento de UAV-DAP a través de cuatro objetivos clave: i) evaluar la efectividad de las características espectrales, geométricas y de vecindad combinadas con técnicas de aprendizaje automático para la clasificación de nubes de puntos e identificación de especies vegetales; ii) desarrollar y validar metodologías para la clasificación de especies de matorral mediterráneo; iii) investigar las relaciones entre las variables geométricas y espectrales derivadas de UAV-DAP y la dinámica de propagación del fuego; iv) y estimar la biomasa aérea (AGB, por sus siglas en inglés) de arbustos y árboles mediante segmentación y clasificación a nivel de planta individual. Al abordar estos objetivos, esta tesis pretende desarrollar metodologías UAV-DAP capaces de proporcionar datos clave sobre la vegetación para mejorar la evaluación de la estructura forestal y el análisis del comportamiento del fuego en entornos mediterráneos., [CA] Els boscos són un dels ecosistemes més crítics de la Terra, cobrint aproximadament el 31% de la superfície terrestre i exercint un paper vital com a sumiders de carboni, reserves de biodiversitat i reguladors del clima. Emmagatzemen al voltant del 45% del carboni terrestre i alberguen la major part de la biodiversitat del planeta, la qual cosa fa que la seua conservació siga essencial per a l'equilibri ecològic. No obstant això, aquests ecosistemes estan cada vegada més amenaçats pels incendis forestals, que compleixen un doble paper en la dinàmica dels boscos: actuen tant com a reguladors naturals com a potencials destructors. La modelització avançada del foc és una eina clau per a millorar la predicció i mitigació dels incendis. Els models d'incendis en 3D, basats en models de dinàmica de fluids computacionals (CFD, per les seues sigles en anglés), requereixen dades d'entrada d'alta resolució sobre la càrrega de combustible, l'estructura de la vegetació i la composició d'espècies. En aquest sentit, un dels majors desafiaments en la modelització d'incendis és l'obtenció de dades precises i d'alta resolució sobre les característiques de la vegetació a escala local. Les tecnologies de teledetecció, en particular la fotogrametria aèria digital amb UAV (UAV-DAP, per les seues sigles en anglés), ofereixen una solució efectiva a nivell de costos i d'alta resolució per a caracteritzar l'estructura forestal. Mitjançant UAV-DAP, és possible obtindre núvols de punts que permeten la segmentació individual de plantes, la classificació d'espècies i l'extracció de característiques geomètriques i espectrals clau per a la dinàmica del foc. Aquestes dades poden utilitzar-se per a estimar variables relacionades amb els incendis. No obstant això, malgrat el seu potencial, UAV-DAP continua estant infrautilitzat a causa de les seues limitacions tècniques inherents i de la falta de metodologies i programari especialitzats que permeten aprofitar plenament tant la seua informació geomètrica tridimensional com el seu contingut espectral. Aquesta tesi té com a objectiu millorar les tècniques de processament d'UAV-DAP a través de quatre objectius clau: i) avaluar l'efectivitat de les característiques espectrals, geomètriques i de veïnat combinades amb tècniques d'aprenentatge automàtic per a la classificació de núvols de punts i la identificació d'espècies vegetals; ii) desenvolupar i validar metodologies per a la classificació d'espècies de matollar mediterrani; iii) investigar les relacions entre les variables geomètriques i espectrals derivades d'UAV-DAP i la dinàmica de propagació del foc; iv) i estimar la biomassa aèria (AGB, per les seues sigles en anglés) d'arbustos i arbres mitjançant segmentació i classificació a nivell de planta individual. En abordar aquests objectius, aquesta tesi pretén desenvolupar metodologies UAV-DAP que permeten proporcionar dades clau sobre la vegetació per a millorar l'avaluació de l'estructura forestal i l'anàlisi del comportament del foc en entorns mediterranis., [EN] Forests are among the most critical ecosystems on Earth, covering approximately 31% of the global land area and serving as vital carbon sinks, biodiversity hotspots, and climate regulators. They store about 45% of terrestrial carbon and support the majority of Earth's biodiversity, making their conservation essential for ecological balance. However, these ecosystems are increasingly threatened by wildfires, which play a dual role in forest dynamics¿acting as both natural regulators and potential destructors. Advancing wildfire modelling is a key tool for improving fire prediction and mitigation efforts. 3D wildfire models, based in computational fluid dynamics (CFD) models, require high-resolution input data on fuel loads, vegetation structure, and species composition. In this regard, one of the greatest challenges in fire modelling is obtaining accurate, high-resolution data on vegetation characteristics at the local scale. Remote sensing technologies, particularly UAV-based Digital Aerial Photogrammetry (UAV-DAP), offer a cost-effective and high-resolution solution for characterize forest structure. Using UAV-DAP is possible to obtain point clouds that enable individual plant segmentation, species classification, and the extraction of key geometric and spectral features relevant to wildfire dynamics. These data can be utilized to estimate fire-related variables. However, despite its potential, UAV-DAP remains underutilized due to inherent technical limitations and the absence of specialized methodologies and software capable of fully leveraging both its three-dimensional geometric and spectral information. This thesis aims to refine UAV-DAP processing techniques through four key objectives: i), evaluating the effectiveness of spectral, geometric, and neighbourhood features combined with machine learning techniques for point cloud classification and vegetation species identification; ii), developing and validating methodologies for classifying Mediterranean shrub species; iii), investigating the relationships between geometric and spectral variables derived from UAV-DAP data and fire spread dynamics; iv), and estimating AGB for shrubs and trees through segmentation and classification at the individual plant level. By addressing these objectives, this thesis intends to develop UAV-DAP methodologies capable of providing key vegetation data to enhance forest structure assessment and fire behaviour analysis in Mediterranean environments., The author of this PhD Thesis gratefully acknowledges the financial support provided by the Spanish Ministerio de Economía, Industria y Competitividad (MCIN/AEI/10.13039/501100011033) and the ESF “Investing in your future.” This support was made possible through the predoctoral contract BES-2017-081920, associated with the FIRMACARTO project (CGL2016-80705-R, 2016-2019), and a subsequent contract under the R+D+i project FireMode (PID2020-117808RB-C21, 2021-2024).

[EN] In view of the increasing intensity and frequency of large forest fires, the study and simulation of fire behavior in Mediterranean forests appears as a crucial task to define prevention and extinction strategies. The physical model Fire Dynamics Simulator (FDS) allows the analysis of fire behavior in 3D scenarios at high levels of resolution by introducing LiDAR point clouds to describe the structure and characteristics of the vegetation. The main objective of this paper is the analysis of fire behavior in Mediterranean forest semi-real scenarios, by modifying structural properties of the forest related to the horizontal and vertical continuity of the vegetation. To achieve this goal, a methodology has been developed to generate scenarios from real segmented and voxelized individuals of Mediterranean shrubland and tree species. These models were obtained from TLS (Terrestrial Laser Scanning) in plots located in La Serra d'Espadà. The methodology considers the distribution of biomass at individual and voxel level, taking into account a heterogeneous distribution of the bulk density of forest fuel, estimated based on the proportion of returns from the point cloud. Finally, for the analysis of the results, the evolution graphs of the following model output variables have been studied: the rate of spread, the heat release rate, the mass loss rate and the percentage of biomass consumed. Preliminary findings indicate that horizontal continuity of vegetation is the main driver for the development of steady and sustained fires. , This work was supported by (i) PID2020-117808RBC21/AEI/10.13039/50110001103 and FSE+ “Invierte en tu Futuro” under Grant PRE2021-099526; and (ii) “Programa de Contratos de Acceso de Personal Investigador Doctor” (PAID-10-22) from “Vicerrectorado de Investigación de la Universitat Politècnica de València”.

[EN] The impact of climate change on vegetation dynamics and wildfire risk has been a subject of considerable research interest. Live fuel moisture content (LFMC) is a critical factor in assessing fire risk and influencing fire ignition and behaviour. Satellite remote sensing techniques provide information on LMFC dynamics, but spatial and temporal resolution hinder understanding in structurally complex forests with interconnected tree and shrub layers. Multi-wavelength terrestrial laser scanning (TLS) sensors can measure the structural and spectral properties of forests and have demonstrated their potential for monitoring LFMC. However, studies of LFMC in shrubs are scarce despite their key role in fire spread. In this study, we investigated the capacity of a dual-wavelength SALCA (Salford Advanced Laser Canopy Analyser) TLS (1063 and 1545 nm) and the single-wavelength Trimble X6 (1500 nm) to estimate LFMC in six Mediterranean forest plots (135 individual plants, 18 species). Analysis at different separate plots and individual-species levels identified key factors affecting LFMC prediction using TLS. At plot level,
linking spectral indices and LFMC is challenging due to species diversity in crown structures, ages, sizes and leaf types. Our results suggest that detector heating by solar radiation could alter the sensor calibration and reduce model accuracy. Nevertheless, in some areas the multiple linear regression models achieved an R2adj up to 0.82 and an RMSE of 7.66%. At the species level, models showed stronger relationships with LFMC (R2adj ranging from 0.43 to 0.88) and a relatively low RMSE (RMSE from 1.92 to 3.97%). Overall, univariate relationships between LFMC and individual-wavelength reflectance were not consistent across species or most plots. Considering these results, combining the capacity of dual TLS devices to estimate LFMC with the structural information that they provide, open a potential to improve field work sampling for wildfire risk assessment. Extending the research to cover a wider range of tree, shrub and herbaceous species in the future will advance our understanding of LFMC dynamics and contribute to more accurate fire behaviour modelling., The author(s) declared that financial support was received for this work and/or its publication. This research is part of the grants
PID2020-117808RB-C21 and PID2024-158591OB-I00, funded by MICIU/AEI/10.13039/501100011033 and the European Social Fund Plus (ESF+). Funding for open access charge: Universitat Politècnica de València (PAID-12-25).

[EN] Accurate, species-specific estimation of aboveground biomass (AGB) at the individual plant level is essential for characterizing forest structure, supporting ecological and wildfire modelling, and enabling fine-scale carbon accounting. This study presents a methodological framework for estimating species-specific AGB at individual plant level in Mediterranean ecosystems using UAV-based digital aerial photogrammetry (UAV-DAP). High-resolution point clouds were processed through a multi-step workflow including object-based segmentation, thirteen species classification and AGB regression modeling. The overall accuracy of species classification across six study areas was 81.6%, with a maximum of 89.9%. The regression models for AGB estimation yielded an average R2 of 0.69 across all species, highlighting species such as Anthyllis cytisoides (R2 = 0.83, RMSE = 0.07 kg, n = 47), Juniperus oxycedrus (R2 = 0.83, RMSE = 3.17 kg, n = 32); or Pinus halepensis (R2 = 0.77, RMSE = 11.79 kg, n = 20). These findings demonstrate the potential of UAV-DAP for practical estimates of AGB. The study underscores UAV-DAP as a cost-effective tool for forest management, ecological monitoring, and biomass assessments, paving the way for broader applications in environmental science and resource management., This research has been supported by the grants BES-2017-081920 and PID2020-117808RB-C21 funded by MCIN/AEI/10.13039/501100011033 and by ESF Investing in your future.

[EN] Modelling fire behaviour in forest fires is based on meteorological, topographical, and vegetation data, including species¿ type. To accurately parameterise these models, an inventory of the area of analysis with the maximum spatial and temporal resolution is required. This study investigated the use of UAV-based digital aerial photogrammetry (UAV-DAP) point clouds to classify tree and shrub species in Mediterranean forests, and this information is key for the correct generation of wildfire models. In July 2020, two test sites located in the Natural Park of Sierra Calderona (eastern Spain) were analysed, registering 1036 vegetation individuals as reference data, corresponding to 11 shrub and one tree species. Meanwhile, photogrammetric flights were carried out over the test sites, using a UAV DJI Inspire 2 equipped with a Micasense RedEdge multispectral camera. Geometrical, spectral, and neighbour-based features were obtained from the resulting point cloud generated. Using these features, points belonging to tree and shrub species were classified using several machine learning methods, i.e., Decision Trees, Extra Trees, Gradient Boosting, Random Forest, and MultiLayer Perceptron. The best results were obtained using Gradient Boosting, with a mean cross-validation accuracy of 81.7% and 91.5% for test sites 1 and 2, respectively. Once the best classifier was selected, classified points were clustered based on their geometry and tested with evaluation data, and overall accuracies of 81.9% and 96.4% were obtained for test sites 1 and 2, respectively. Results showed that the use of UAV-DAP allows the classification of Mediterranean tree and shrub species. This technique opens a wide range of possibilities, including the identification of species as a first step for further extraction of structure and fuel variables as input for wildfire behaviour models., Grants BES-2017-081920 and PID2020-117808RB-C21 funded by MCIN/AEI/10.13039/5011
00011033 and by ESF Investing in your future.

This dataset contains RGB and multispectral point clouds of the Sierra Calderona Natural Park, obtained by UAV digital aerial photogrammetry., Grants BES-2017-081920 and PID2020-117808RB-C21 funded by MCIN/AEI/10.13039/501100011033 and by ESF Investing in your future.

[EN] Recognizing the species composition of an ecosystem is essential for conservation and land management. This study presents the software Class3Dp, a supervised classifier of vegetation species for coloured point clouds. Class3Dp is run through a graphical user interface (GUI) that allows for the selection of training samples from RGB or MS (multispectral) clouds and their classification based on geometric, spectral and neighbourhood features, along with different machine learning methods, obtaining the point cloud classified according to the classes (species) introduced. A case study is shown where a classification of ground and vegetation is carried out, obtaining an overall accuracy (OA) of 0.94 in the RGB classification and 0.95 in the MS. Points classified as vegetation were re-classified in the species Anthyllis cytisoides L., Chamaerops humilis L., Cistus monspeliensis L., Pistacia lentiscus L. and Quercus coccifera L., obtaining an OA of 0.86 in the RGB classification and 0.87 in the MS., This research has been supported by the grants BES-2017-081920 and PID2020-117808RB-C21 funded by MCIN/AEI/10.13039/501100011033 and by ESF Investing in your future. Funding for open access charge: Universitat Politècnica de València.

[EN] Class3Dp is a software tool developed for supervised classification of photogrammetric point clouds. It allows the selection of training samples and classification of point clouds according to different features and machine learning models. Point clouds obtained from photogrammetric algorithms (e.g., Structure from Motion) include spectral information recorded in the image acquisition process using RGB, multispectral or hyperspectral sensors. The software extracts spectral and geometric features from the point clouds to be used for point classification., The financial support for its development has been provided by the grant BES-2017-081920 and the R&D&I projects CGL2016-80705-R and PID2020-117808RB-C21 funded by MCIN/AEI/10.13039/501100011033 and by "FSE Invierte en tu futuro".

[EN] This article presents a methodology to estimate the live fuel moisture content (LFMC), a critical factor in the spread of forest fires, through machine learning tools. Random forest models were trained using field LFMC samples collected bi-weekly for 18 consecutive months in 43 shrubland plots in the Valencian region, a Mediterranean zone in eastern Spain. LFMC predictions were obtained for the weighted average of LFMC values, calculated using the Fraction of Canopy Cover (FCC) of dominant species as weights. Furthermore, a specific model was defined for predicting LFMC for the Rosmarinus officinalis species. A Forward Feature Selection (FFS) with a Leave-Location-Out Cross Validation (LLOCV) method was used to select predictors extracted from a spatiotemporal data set, which includes different spectral indices obtained from Sentinel-2 imagery and meteorological variables obtained from measurements at weather stations, along with other seasonal, geographical or topographic variables. Model predictions were validated with a LLOCV procedure, and also using independent field measurements of LFMC in another period with changes in the precipitation regime and average temperatures. Variables selected by FFS for the two LFMC models were: the cumulative precipitation in the previous 60
days (p60), the average of the daily mean temperature in the previous 60 days (t60), together with the Y-UTM coordinate and the sine and cosine of the day of the year. LFMC predictions for the weighted average of LFMC values also introduced the Transformed Chlorophyll Absorption Ratio Index (TCARI), resulting in an R2 of 68.1%. However, LFMC for the Rosmarinus officinalis species used the ratio between TCARI and the Optimized SoilAdjusted Vegetation Index (OSAVI), in addition to the average daily minimum relative humidity in the 15 days prior to the date considered (R2 = 74.9 %). LFMC time series analysis showed that the general trend of LFMC measures is satisfactorily captured by the predictions. Spatial and temporal variations in LFMC were analyzed throughout thematic maps in the studied area during the wildfire season., Ma Alicia Arcos gives thanks for the help received by the Universitat Politecnica de Valencia through a pre-doctoral contract financed in the call, PAID-01-19, subprogram 1, and the open access charge: Universitat Politecnica de Valencia (PAID-12-23) . This research was funded by through a collaboration agreement between the company Red Electrica de Espana S.A.U. and the Universitat Politecnica de Valencia (2020-2023) , as well as the research project PID2020-117808RB-C21 funded by MCIN/AEI/10.13039/501100011033 and by ESF Investing in your future.

[EN] Incorporating explicit spatial heterogeneity in fuel characterization will reduce uncertainty in fire modelling, especially in Mediterranean forests which are characterized by complex structures. We provide a reproducible workflow to integrate bulk density estimates derived from terrestrial laser scanner (TLS) data into Fire Dynamics Simulator (FDS), and evaluate fire behavior sensitivity to wind speed and fuel moisture content (FMC) in two Mediterranean forest plots. The methodology incorporates the 3D fuel distribution into FDS derived from point clouds. 210 simulations were conducted by modifying wind speed and vegetation FMC. Results confirm the strong negative effect of FMC on fire behavior, the positive effect of wind on fire propagation, and evidencing the importance of their combined effect. Fire behavior sensitivity proved to be strongly scenario-dependent, demonstrating that structural heterogeneity plays a critical role with direct implications for assessing fire propagation probability under varying moisture and wind regimes in real forest stands., This research was funded by grant PRE2021-099526 and projects
I+D+I PID2020-117808RB-C21 and PID2024-158591OB-I00, funded
by MCIN/AEI/ 10.13039/ 501100011033 and FSE+. Funding for open
access charge: Universitat Politecnica de Valencia.

[EN] This paper presents empirical models developed through stepwise multiple linear regression to estimate the live fuel moisture content (LFMC) in a Mediterranean area. The models are based on LFMC data measured in 50 field plots, considering four groups with similar bioclimatic characteristics and vegetation types (trees and shrubs). We also applied a species-specific LFMC model for Rosmarinus officinalis in plots with this dominant species. Spectral indices extracted from Sentinel-2 images and their averages over the study time period in each plot with a spatial resolution of 10 m were used as predictors, together with interpolated meteorological, topographic, and seasonal variables. The models achieved adjusted R2 values ranging between 52.1% and 74.4%. Spatial and temporal variations of LFMC in shrub areas were represented on a map. The results highlight the feasibility of developing satellite-derived LFMC operational empirical models in areas with various vegetation types and taking into account bioclimatic zones. The adjustment of data through GAM
(generalized additive models) is also addressed in this study. The different error metrics obtained reflect that these models provided a better fit (most adjusted R2 values ranged between 65% and 74.1%) than the linear models, due to GAMs being more versatile and suitable for addressing complex problems such as LFMC behavior., Ma Alicia Arcos gives thanks for the help received by the Universitat Politecnica de Valencia through a pre-doctoral contract financed in the call, PAID-01-19, subprogram 1. This research was funded by through a collaboration agreement between the company Red Electrica de Espana S.A.U. and the Universitat Politecnica de Valencia (2020-2023), as well as the research project PID2020-117808RB-C21 funded by MCIN/AEI/10.13039/501100011033 and by ESF Investing in your future.

[EN] Unmanned aerial vehicles (UAVs) equipped with RGB, multispectral, or thermal cameras have demonstrated their potential to provide high-resolution data before, during, and after wildfires and prescribed burns. Pre-burn point clouds generated through the photogrammetric processing of UAV images contain geometrical and spectral information of vegetation, while active fire imagery allows for deriving fire behavior metrics. This paper focuses on characterizing the relationship between the fire rate of spread (RoS) in prescribed burns and a set of independent geometrical, spectral, and neighborhood variables extracted from UAV-derived point clouds. For this purpose, different flights were performed before and during the prescribed burning in seven grasslands and open forest plots. Variables extracted from the point cloud were interpolated to a grid, which was sized according to the RoS semivariogram. Random Forest regressions were applied, obtaining up to 0.56 of R2 in the different plots studied. Geometric variables from the point clouds, such as planarity and the spectral normalized blue¿red difference index (NBRDI), are related to fire RoS. In analyzing the results, the minimum value of the eigenentropy (Eigenentropy_MIN), the mean value of the planarity (Planarity_MEAN), and percentile 75 of the NBRDI (NBRDI_P75) obtained the highest feature importance. Plot-specific analyses unveiled distinct combinations of geometric and spectral features, although certain features, such as Planarity_MEAN and the mean value of the grid obtained from the standard deviation of the distance between points (Dist_std_MEAN), consistently held high importance across all plots. The relationships between pre-burning UAV data and fire RoS can complement meteorological and topographic variables, enhancing wildfire and prescribed burn models., This research was primarily supported by grants BES-2017-081920 and PID2020-117808RBC21 from MCIN/AEI/10.13039/501100011033 and ESF Investing in your future. Additional support
was provided by the US National Science Foundation under Grant No. OIA-2119689 and the Strategic
Environmental Research and Development Program (SERDP), grant number RC20-1025.