Satellite images enable the analysis of the spatial and temporal distribution of burned areas (BA) and burn severity (BS) to quantify the impacts of wildfires. However, the generation of reliable maps requires algorithms and tools available to the user for regional to global analyses. We present a public QGIS plugin (BAD, Burned Area Detector) for automatic BA and BS mapping using pre- and post-fire Sentinel-2 multispectral images. The plugin also incorporates a validation module for assessing the accuracy of output maps. The BA detection is based on a multi-criteria soft computing approach that incorporates experts’ fuzzy knowledge and its integration, combined with a region growing algorithm. BS is estimated using the difference of the Normalized Burn Ratio (NBR) index.The plugin was tested on wildfire events in Spain (summer 2022) and California (winter 2025). Besides proving the functionalities of BAD, these test cases confirm the robustness of the algorithm when applied automatically to Mediterranean regions and its flexibility in ingesting different data sources (active fires as seeds for the region growing). Results across all studied areas show an average omission error of 9.57%, a commission error of 16.56%, and a Dice coefficient of 89.62%.

Accurate burned area (BA) mapping is essential for assessing wildfire impacts on ecosystems and climate. While existing BA products derived from coarse-resolution sensors (e.g., MODIS) have primarily relied on Landsatbased validation protocols, the advent of higher-resolution products such as those from Sentinel-2 necessitates adapted validation methodologies to match their enhanced spatial detail. This study presents the Sentinel-2 Burned Area Validation Grid (S2BAVG); a global sampling framework designed to support BA validation using Sentinel-2 imagery. S2BAVG consists of 19,263 non-overlapping spatial units covering the global land surface, addressing limitations of previous grids by ensuring full orbital coverage and eliminating overlaps, thereby enabling consistent sampling and rigorous validation. Key attributes-including fire activity indicators and cloud-free image availability- facilitate the implementation of stratified sampling designs. Additionally, we provide an open-source framework to support the sampling process with customizable input parameters. The framework includes statistical inference tools to estimate accuracy metrics and their standard errors, ensuring rigorous BA product assessment. By leveraging Sentinel-2 ' s high spatial and temporal resolution, S2BAVG provides a flexible and standardized methodology for BA validation. The S2BAVG tile grid dataset and sampling framework (with an illustrative sampling design approach) are openly available at https://github.com/magi franquesa/S2BAVG, promoting reproducibility and enabling broader applications in fire science and Earth observation.

This work focuses on monitoring wildfires using remote sensing and time series change detection of multi-spectral and Synthetic Aperture Radar (SAR) data. Remote sensing can offer detailed information on fire conditions and risks paired with frequent revisit times. Multispectral satellites, like Landsat and Sentinel-2 (S-2), data have been, and still are, widely used in Earth Observation thanks to the wide range of wavelengths and the frequent rate of observation. On the other hand, Sentinel-1 (S-1) SAR data offers the advantage of independence from solar illumination and weather conditions. This work focuses on generating and classifying time series of Landsat and Sentinel-2 data into burned/unburned areas using a random forest (RF) algorithm. Python and Google Earth Engine (GEE) tools were developed to automate the extraction of fire reference perimeters from multi-spectral images and merging the results to form a burned area dataset, while providing support to the user. S-2 burned area perimeters over the Sahel (Africa) and Amazon (South America) regions were used to assess the sensitivity to burned surfaces of S-1 SAR data pre- and post-fire backscatter in the VV and VH polarizations. Statistical tests and data visualization were used to assess the change in intensity of the backscatter signal (γ0) after a fire event. The changes in the distribution of γ0 values were found significant for the majority of the test cases, and more visible in the VH polarization mode. Classification tests were conducted using the S-1 backscatter values and dedicated radar indices to identify the occurrence of fires, considering different time configurations. Longer temporal baselines of S-1 acquisitions were found to produce more accurate results in detection in high fire activity regions. Although S-1 SAR shows promise in wildfire detection, challenges remain in interpreting radar returns. Further research could explore other polarization methods and longer temporal baselines to enhance accuracy in different biomes.

Thermal InfraRed (TIR) Remote Sensing is a well-consolidated approach to detect ground thermal anomalies for geological, environmental and urban scenarios. Specifically, several methodologies have been developed for the TIR imagery analysis to retrieve the Land Surface Temperature (LST) and describe the thermal state of the Earth’s surface. In volcanic frameworks, the analysis of LST time series represents a valid tool for a fast characterization of the shallow thermal field, supporting the surveillance networks in monitoring their status, specifically for the areas inaccessible because of the high volcanic hazard. Here, we propose a workflow to detect the thermal patterns in volcanic areas by analyzing time series of satellite TIR images using the Independent Component Analysis (ICA) technique. In particular, the first step of the workflow relies on the retrieval of LST time series from Landsat-8 (L8) TIR nighttime acquisitions, which have spatial and temporal resolutions equal to 100 m and 16 days, respectively, acceptable for our purposes. We selected the nighttime images because they allow us to reduce the exogenous effects, as well as those related to the solar radiation. Therefore, we estimate the LST parameter by considering the Radiative Transfer Equation (RTE) based on the use of a single thermal band, as long as having the surface emissivity and the atmospheric information about the investigated area. The second step of the considered workflow deals with the application of the ICA method to the retrieved LST time series to identify the statistically independent components (ICs) of the LST multivariate dataset. We verify the robustness of the proposed workflow by analyzing the volcanic site of Campi Flegrei caldera (Southern Italy), which represents a well-suitable case study for the occurrence of several endogenous and exogenous phenomena. We first achieved the 2013 – 2022 LST time series and subsequently analyzed the four components identified by the ICA. We compare these main thermal patterns with other available independent datasets, for example, the seismicity, the ground deformation field and the depth of the water table in the area, proving: (i) the existence of a positive thermal anomaly at the Solfatara crater with endogenous nature; (ii) the occurrence of exogenous processes at the Agnano plain; (iii) the existence of peculiar climatic pattern at the Astroni crater. In conclusion, we remark that the proposed methodology allows the identification of the nature of thermal anomalies, even for complex volcanic scenarios where several processes of different nature occur interfering with each other.

Satellite data provide the spatial distributions of burned areas worldwide; assessing their accuracy and comparing burned area estimates from different products is relevant to gain insights into their reliability and sources of error. We compared BA maps derived from multispectral satellite data with different spatial resolutions, ranging from Planet (3 m) to Sentinel-2 (S2, 10–20 m), Sentinel-3 (S3, 300 m), and MODIS (250–500 m), over selected African sites for the year 2019. Planet and S2 images were processed to derive BA maps with a supervised Random Forest algorithm and used to assess the spatial agreement of the FireCCISFD20, FireCCI51, FireCCIS311, and MCD64A1 products by computing omission and commission errors, Dice Coefficient, and Relative bias. The products based on S2 images showed the greatest agreement with the very high-resolution Planet BA maps (overall Dice Coefficient was found to be greater than 80%). The coarse-resolution products showed a lower spatial agreement with reference perimeters. Among the coarse spatial resolution products, FireCCIS311 was found to outperform the others. The spatial resolution of satellite data was found to be influential on accuracy, with the omission error greater than the commission (RelB < 0) for coarser resolution BA products. The spatial patterns of burns and the vegetation type were found to be significant in the mapping accuracy, and BA detection in Sahelian savannas was found to be more accurate. This study provides insights into the variability of the spatial accuracy of different burned area products derived from very high- to coarse-resolution satellite imagery.

In volcanic regions, the analysis of Thermal InfraRed (TIR) satellite imagery for Land Surface Temperature (LST) retrieval is a valid technique to detect ground thermal anomalies. This allows us to achieve rapid characterization of the shallow thermal field, supporting ground surveillance networks in monitoring volcanic activity. However, surface temperature can be influenced by processes of different natures, which interact and mutually interfere, making it challenging to interpret the spatio-temporal variations in the LST parameter. In this paper, we use a workflow to detect the main thermal patterns in active volcanic areas by analyzing the Independent Component Analysis (ICA) results applied to satellite nighttime TIR imagery time series. We employed the proposed approach to study the surface temperature distribution at the Campi Flegrei caldera volcanic site (Southern Italy, Naples) during the 2013–2022 time interval. The results revealed the contribution of four main distinctive thermal patterns, which reflect the endogenous processes occurring at the Solfatara crater, the environmental processes affecting the Agnano plain, the unique microclimate of the Astroni crater, and the morphoclimatic aspects of the entire volcanic area.

A study was carried out to investigate the effects of wildfires on lake water quality using a source dataset of 2024 lakes worldwide, covering different lake types and ecological settings. Satellite-derived datasets (Lakes_cci and Fire_cci) were used and a Source Pathway Receptor approach applied which was conceptually represented by fires (burned area) as a source, precipitation/drought representing transport dynamics, and lakes as the ultimate receptor. This identified 106 lakes worldwide that are likely prone to be impacted by wildfires via a terrestrial pathway. Satellite-derived chlorophyll-a (Chl-a) and turbidity variables were used as indicators to detect changes in lake water quality potentially induced by wildfires over a four-year period. The lakes with the largest catchment areas burned and characterized by regular annual fires were located in Africa. Evidence for a strong influence of wildfires was not found across the dataset examined, although clearer responses were seen for some individual lakes. However, among the hydro-morphological characteristics examined, lake depth was found to be significant in determining Chl-a concentration peaks which were higher in shallow and lower in deep lakes. Lake turbidity responses indicated a dependence on lake catchment and weather conditions. While wildfires are likely to contribute to the nutrient load of lakes as found in previous studies, it is possible that in many cases it is not a dominant pressure and that its manifestation as a signal in lake Chl-a or turbidity values depends to a large part on lake typology and catchment characteristics. Assessment of lake water quality changes six months after a fire showed that Chl-a concentrations either increased, decreased, or showed no changes in a similar number of lakes, indicating that a lake specific ecological and hydro-morphological context is important for understanding lake responses to wildfires.

Lakes have been observed as sentinels of climate change. In the last decades, global warming and increasing aridity has led to an increase in both the number and severity of wildfires. This has a negative impact on lake catchments by reducing forest cover and triggering cascading effects in freshwater ecosystems. In this work we used satellite remote sensing to analyse potential fire effects on lake water quality of Lake Baikal (Russia), considering the role of runoff and sediment transport, a less studied pathway compared to fire emissions transport. The main objectives of this study were to analyse time series and investigate relationships among fires (i.e., burned area), meteo-climatic parameters and water quality variables (chlorophyll-a, turbidity) for the period 2003–2020. Because Lake Baikal is oligotrophic, we expected detectable changes in water quality variables at selected areas near the three mains tributaries (Upper Angara, Barguzin, Selenga) due to river transport of fire-derived burned material and nutrients. Time series analysis showed seasonal (from April to June) and inter-annual fire occurrence, precipitation patterns (high intensity in summer) and no significant temporal changes for water quality variables during the studied periods. The most severe wildfires occurred in 2003 with the highest burned area detected (36,767 km2). The three lake sub-basins investigated have shown to respond differently according to their morphology, land cover types and meteo-climatic conditions, indicating their importance in determining the response of water variables to the impact of fires. Overall, our finding suggests that Lake Baikal shows resilience in the medium-long term to potential effects of fires and climate change in the region.

Coarse resolution sensors are not very sensitive at detecting small fire patches, making current estimations of global burned areas (BA) very conservative. Using medium or high-resolution sensors to generate BA products becomes then a priority, particularly in areas where fires tend to be small and frequent. Building on previous work that developed a small fire dataset (SFD) for Sub-Saharan Africa for 2016, this paper presents a new version of the dataset for 2019 using the two Sentinel-2 satellites (A and B) and VIIRS active fires. Total estimated BA was 4.8 Mkm2. This value was much higher than estimations from two global, coarser-spatial resolution BA products based on MODIS data for the same area and period: 80 % greater than estimates from FireCCI51 (based on MODIS 250 m bands) and 120 % larger than MCD64A1 (based on MODIS 500 m bands). The main differences were observed in those months with higher fire occurrence (November to January for the Northern Hemisphere regions and June to September for the Southern Hemisphere ones). Accuracy assessment of the SFD product was based on a novel sampling strategy designed to obtain independent fire reference perimeters. Validation results showed remarkable high accuracy values comparing to existing global BA products. Overall omission errors (OE) were estimated as 8.5 %, commission errors (CE) as 15.0 %, with a Dice Coefficient of 87.7 %. All of these estimations implied significant improvements over the global, coarser spatial resolution BA products (OE > 50 % and CE > 20 % for the same area and period), as well as over the previous SFD product for 2016 of the same area, generated from a single Sentinel-2 satellite and MODIS active fires (OE = 26.5 % and CE = 19.3 %). Temporal accuracies greatly increased as well with the new product, with 92.5 % of fires detected within the first 10 days of occurrence.

Fires devastated Europe during the summer of 2021, with hundreds of events burning across the Mediterranean, causing unprecedented damage to people, properties, and ecosystems [...]

The paper proposes a multi-criteria and data driven fusion approach whose semantics can be explained in terms of attitude towards decisions. It is exemplified to assess environmental status from remote sensing images in order to identify hot spot of critical situations and anomalies induced by wildfires, floods, desertification, erosion etc. by fusing multiple factors defined by experts knowledge. The fusion function is an Ordered Weighted Averaging (OWA) operator, whose behaviour is here characterized by degrees of pessimism and democracy. The paper proposes to explain the semantics of the fusion function learnt from few ground truth data available, i.e., the OWA operator, by computing its degrees of pessimism/optimism and democracy/monarchy, which are defined as semantic interpretations of both orness and dispersions respectively. Pessimism indicates if the fused map is more prone to commission (overestimation) or omission (underestimation) errors, while democracy indicates how many factors contribute to the generation of the map. The approach is exemplified to map the flooded areas from remote sensing by considering different models based on distinct spectral indexes and domain experts.

The availability of high-resolution reference datasets representing in space and time and with high accuracy areas affected by fires is strategic for the validation of remotely-sensed Burned Area (BA) products. This paper proposes a methodology designed to build a burned area reference dataset from Sentinel-2 (S2) images at continental scale by implementing a stratified random sampling scheme. Representative sample units are selected across biomes and regions with high/low fire activity; each unit covers the extent of a S2 tile (similar to 10 000 km(2)) where image time series are classified with a supervised Random Forest algorithm to extract fire perimeters by exploiting visible to near and short-wave infrared S2 wavebands at 10 to 20 m spatial resolution. Time series have to satisfy requirements on maximum cloud cover, maximum time interval between consecutive images and minimum length to be suitable for being selected and processed. The proposed methodology was applied to Sub-Saharan Africa for the year 2019 to select 50 S2 sample units where time series were processed to deliver fire reference perimeters for accuracy assessment of regional BA products. Average series length is 140 days with the longest series in the savanna biome (maximum length is 355 days, 29 consecutive S2 images) and a total of 695 S2 images were processed to build the 2019 reference dataset. This dataset was compared to burned areas derived from very-high resolution Planetscope images over five S2 tiles obtaining 15.5% omission and 11.6% commission errors. To exemplify the use of this reference dataset, S2 perimeters were used to validate the NASA MCD64A1 Collection 6 and the ESA FireCCI51 BA products. The reference dataset has been added to the Burned Area Reference Database (BARD) (Franquesa et al., 2020) and is publicly available at https://doi.org/10.21950/VKFLCH.

Thermal features of environmental systems are increasingly investigated after the development of remote sensing technologies; the increasing availability of Earth Observation (EO) missions allows the retrieval of the Land Surface Temperature (LST) parameter, which is widely used for a large variety of applications (Galve et al., 2018). In volcanic environment, the LST is an indicator of the spatial distribution of thermal anomalies at the ground surface, supporting designed tools for monitoring purposes (Caputo et al., 2019); therefore, LST can be used to understand endogenous processes and to model thermal sources. In this framework, we present the results of activities carried out in the FLUIDs PRIN project, which aims at the characterization and modeling of fluids migration at different scales (https://www.prinfluids.it/). We propose a multi-scale analysis of thermal data at Campi Flegrei caldera (CFc); this area is well known for hosting thermal processes related to both magmatic and hydrothermal systems (Chiodini et al., 2015; Castaldo et al., 2021). Accordingly, data collected at different scales are suitable to search out local thermal trends with respect to regional ones. In particular, in this work we compare LST estimated from Landsat satellite images covering the entire volcanic area and ground measurements nearby the Solfatara crater. Firstly, we exploit Landsat data to derive time series of LST by applying an algorithm based on Radiative Transfer Equations (RTE) (Qin et al., 2001; Jimenez-Munoz et al., 2014). The algorithm exploits both thermal infrared (TIR) and visible/near infrared (VIS/NIR) bands of different Landsat missions in the period 2000-2021; we used time series imagery from Landsat 5 (L5), Landsat 7 (L7) and Landsat 8 (L8) satellite missions to retrieve the thermal patterns of the CFc area with spatial resolutions of 30 m for VIS/NIR bands and 60 m to 120 m for TIR bands. Theoretical frequency of acquisition of the Landsat missions is 16 days that is reduced over the study area by cloud cover: Landsat images with high cloud cover were in fact discarded from the time series. In particular, we process both the daily acquisitions as well nighttime data to provide thermal features at the ground surface in the absence of solar radiation. To emphasize the thermal anomalies of endogenous phenomena, the retrieved LST time-series are corrected following these steps: (i) removal of spatial and temporal outliers; (ii) correction for adiabatic gradient of the air with the altitude; (iii) detection and removal of the seasonal component. Regarding to the ground-based acquisitions, we consider the data collected by the Osservatorio Vesuviano, National Institute of Geophysics and Volcanology (OV- INGV, Italy, Naples); the dataset consists of 151 thermal measurements distributed within the 2004-2021 time-interval and acquired inside the Solfatara and Pisciarelli areas at a depth of 0.01 m below the ground surface. Similarly, we process this dataset following corrections (i) and (iii). Finally, we compare the temporal evolution of thermal patterns retrieved by the satellite and ground-based measurements, highlighting the supporting information provided by LST and its integration with data at ground.

Land surface temperature (LST) is a manifestation of the surface thermal environment (LSTE) and an important driver of physical processes of surface land energy balance at local to global scales. Tenerife is one of the most heterogeneous islands among the Canaries from a climatological and bio-geographical point of view. We study the surface thermal conditions of the volcanic island with remote sensing techniques. In particular, we consider a time series of Landsat 8 (L8) level 2A images for the period 2013 to 2019 to estimate LST from surface reflectance (SR) and brightness Temperature (BT) images. A total of 26 L8 dates were selected based on cloud cover information from metadata (land cloud cover < 10%) to estimate pixel-level LST with an algorithm based on Radiative Transfer Equations (RTE). The algorithm relies on the Normalized Difference Vegetation Index (NDVI) for estimating emissivity pixel by pixel. We apply the Independent Component Analysis (ICA) that revealed to be a powerful tool for data mining and, in particular, to separate multivariate LST dataset into a finite number of components, which have the maximum relative statistical independence. The ICA allowed separating the land surface temperature time series of Tenerife into 11 components that can be associated with geographic and bioclimatic zones of the island. The first ten components are related to physical factors, the 11th component, on the contrary, presented a more complex pattern resulting possibly from its small amplitude and the combination of various factors into a single component. The signal components recognized with the ICA technique, especially in areas of active volcanism, could be the basis for the space-time monitoring of the endogenous component of the LST due to surface hydrothermal and/or geothermal activity. Results are encouraging, although the 16-day revisit frequency of Landsat reduces the frequency of observation that could be increased by applying techniques of data fusion of medium and coarse spatial resolution images. The use of such systems for automatic processing and analysis of thermal images may in the future be a fundamental tool for the surveillance of the background activity of active and dormant volcanoes worldwide.

The 2021 European summer season has been one of the most intense for Mediterranean regions that experienced a heatwave in August, determining the onset of several fire events across regions in southern Europe [...]

Sentinel-2 (S2) multi-spectral instrument (MSI) images are used in an automated approach built on fuzzy set theory and a region growing (RG) algorithm to identify areas affected by fires in Mediterranean regions. S2 spectral bands and their post- and pre-fire date (Δpost-pre) difference are interpreted as evidence of burn through soft constraints of membership functions defined from statistics of burned/unburned training regions; evidence of burn brought by the S2 spectral bands (partial evidence) is integrated using ordered weighted averaging (OWA) operators that provide synthetic score layers of likelihood of burn (global evidence of burn) that are combined in an RG algorithm. The algorithm is defined over a training site located in Italy, Vesuvius National Park, where membership functions are defined and OWA and RG algorithms are first tested. Over this site, validation is carried out by comparison with reference fire perimeters derived from supervised classification of very high-resolution (VHR) PlanetScope images leading to more than satisfactory results with Dice coefficient > 0.84, commission error < 0.22 and omission error < 0.15. The algorithm is tested for exportability over five sites in Portugal (1), Spain (2) and Greece (2) to evaluate the performance by comparison with fire reference perimeters derived from the Copernicus Emergency Management Service (EMS) database. In these sites, we estimate commission error < 0.15, omission error < 0.1 and Dice coefficient > 0.9 with accuracy in some cases greater than values obtained in the training site. Regression analysis confirmed the satisfactory accuracy levels achieved over all sites. The algorithm proposed offers the advantages of being least dependent on a priori/supervised selection for input bands (by building on the integration of redundant partial burn evidence) and for criteria/threshold to obtain segmentation into burned/unburned areas.

The paper proposes a fully automatic algorithm approach to map burned areas from remote sensing characterized by human interpretable mapping criteria and explainable results. This approach is partially knowledge-driven and partially data-driven. It exploits active fire points to train the fusion function of factors deemed influential in determining the evidence of burned conditions from reflectance values of multispectral Sentinel-2 (S2) data. The fusion function is used to compute a map of seeds (burned pixels) that are adaptively expanded by applying a Region Growing (RG) algorithm to generate the final burned area map. The fusion function is an Ordered Weighted Averaging (OWA) operator, learnt through the application of a machine learning (ML) algorithm from a set of highly reliable fire points. Its semantics are characterized by two measures, the degrees of pessimism/optimism and democracy/monarchy. The former allows the prediction of the results of the fusion as affected by more false positives (commission errors) than false negatives (omission errors) in the case of pessimism, or vice versa; the latter foresees if there are only a few highly influential factors or many low influential ones that determine the result. The prediction on the degree of pessimism/optimism allows the expansion of the seeds to be appropriately tuned by selecting the most suited growing layer for the RG algorithm thus adapting the algorithm to the context. The paper illustrates the application of the automatic method in four study areas in southern Europe to map burned areas for the 2017 fire season. Thematic accuracy at each site was assessed by comparison to reference perimeters to prove the adaptability of the approach to the context; estimated average accuracy metrics are omission error = 0.057, commission error = 0.068, Dice coefficient = 0.94 and relative bias = 0.0046.

This work describes an approach for building a dataset of fire reference perimeters over sub-Saharan Africa (latitude range 25 degrees N-35 degrees S) from Sentinel-2 (S2) time series based on a sampling scheme designed on the characteristics of the S2 tiling system. A stratified random sampling scheme is adopted to select 50 validation sites distributed over the African continent and across Olson biomes (Olson et al. 2001) and fire intensity strata. S2 time series are extracted over each validation site after a preliminary analysis of data availability; S2 consecutive pairs are classified with a Random Forest (RF) algorithm for building a dataset of fire reference perimeters.