Search

The formation of forest fire burned area, influenced by a variety of factors such as meteorology, topography, vegetation, and human intervention, is a dynamic process of fire line burning that develops from the point of ignition to the boundary of the burned area. Accurately simulating and predicting this dynamic process can provide a scientific basis for forest fire control and suppression decisions. In this study, five typical forest fires located in different regions of China were used as the study object. The straight path distances from the ignition point grid to each grid on fire line in Sentinel-2 imageries for each forest fire were used as the target variables. We obtained the values of 11 independent variables for each pathway, including wind speed component, Temperature, Relative Humidity, Elevation, Slope, Aspect, Degree of Relief, Normalized Difference Vegetation Index, Vegetation Type, Fire Duration, and Gross Domestic Product reflecting human intervention capacity for fires. The value of each target variable and that of its corresponding independent variable constituted a sample. Four machine learning models, such as Random Forest (RF), Gradient Boosting Decision Trees (GBDT), Support Vector Machine (SVM), and Multilayer Perceptron (MLP), were trained using 80% effective samples from four forest fires, and 20% used to verify the above models. The hyper-parameters of each model were optimized using grid search method. After analyzing the validation results of models which showed temperature as a non-significant variable, the training and validation process of models above was repeated after excluding temperature. The results show that RF is the optimal model with 49.55 m for root mean square error (RMSE), 29.19 m for mean absolute error (MAE) and 0.9823 for coefficient of determination (R2). This study used the RF model to construct the shape of burned areas by predicting lengths of all straight path distances from the ignition point to the fire line. The study can dynamically capture the development of forest fire scenes.

This study uses a combined research approach based on remote-sensing and numerical modeling to quantify the effects of burned areas on the surface climate in the two Brazilian biomes most affected by fires: the tropical savanna and the Amazon rainforest. Our estimates indicate that between 2007 and 2020, approximately 6% of the savanna and 2% of the rainforest were burned on average. Non-parametric regressions based on 14-year climate model simulations indicate that latent heat flux decreases on average by approximately 0.17 W m−2 in the savanna and 0.60 W m−2 in the rainforest per each 1 km2 burned, with most of the impacts registered during the onset of the wet season. Sensible and ground heat fluxes are also impacted but at less intensity. Surface air is also warmer and drier, especially over rainforest burned sites. On average, fire reduced gross primary production in the savanna and rainforest by 12% and 10%, respectively, in our experiments.

Forest fires have become a major concern in the northern parts of Morocco, particularly in the Tangier-Tetouan-Al Hoceima (TTA) region, causing significant damage to the environment and human lives. To address this pressing issue, this study proposes an approach that utilizes remote sensing (RS) and machine learning (ML) techniques to detect burned areas in the TTA region within the Google Earth Engine platform. The study focuses on burned areas resulting from forest fires in three specific locations in the TTA region that have experienced such fires in recent years, namely Tangier-Assilah in 2017, M’diq Fnideq in 2020, and Chefchaouen in 2021. In our study, we extensively explored multiple combinations of spectral indices, such as normalized burn ratio (dNBR), normalized difference vegetation index (dNDVI), soil-adjusted vegetation index (dSAVI), and burned area index (dBAI), in conjunction with Sentinel-2 (S2) satellite images. These combinations were employed within the Random Forest (RF) algorithm, allowing us to draw important conclusions. Initially, we assess the individual effectiveness of the dNBR index, which yields accuracy rates of 83%, 90%, and 82% for Tangier-Assilah, Chefchaouen, and M’diq Fnideq, respectively. Recognizing the need for improved outcomes, we expand our analysis by incorporating spectral indices and S2 bands. However, the results obtained from this expanded combination lack consistency and stability across different locations. While Tangier-Assilah and M’diq Fnideq experience accuracy improvements, reaching 95% and 88%, respectively, the inclusion of Sentinel bands has an adverse effect on Chefchaouen, resulting in a decreased accuracy of 87%. To achieve optimal accuracy, our focus shifted towards the combination of dNBR and the other spectral indices. The results were truly remarkable, with accuracy rates of 96%, 97%, and 97% achieved for Tangier-Assilah, Chefchaouen, and M’diq Fnideq, respectively. Our decision to prioritize the spectral indices was based on the feature importance method, which highlights the significance of each feature in the classification process. The practical implications of our study extend to fire management and prevention in the TTA region. The insights gained from our analysis can inform the development of effective policies and strategies to mitigate the impact of forest fires. By harnessing the potential of RS and ML techniques, along with the utilization of spectral indices, we pave the way for enhanced fire monitoring and response capabilities in the region.

Protected areas (PA) play an important role in minimizing the effects of soil erosion in watersheds. This study evaluated the performance of machine learning models, specifically support vector machine with linear kernel (SVMLinear), support vector machine with polynomial kernel (SVMPoly), and random forest (RF), on identifying indicators of soil erosion in 761 sub-watersheds and PA in northern Portugal, by using soil erosion by water in Europe, according to the revised universal soil loss equation (RUSLE2015), as target variable. The parameters analyzed were: soil erosion by water in Europe according to the revised universal soil loss equation (RUSLE2015), total burned area of the sub-watershed in the period of 1975-2020, fire recurrence, topographic wetness index (TWI), and the morphometric factors, namely area (A), perimeter (P), length (L), width (W), orientation (O), elongation ratio (Re), circularity ratio (Rc), compactness coefficient (Cc), form factor (Ff), shape factor (Sf), DEM, slope, and curvature. The median coefficient of determination (R2) for each model was RF (0.61), SVMpoly (0.68), and SVMLinear (0.54). Regarding the analyzed parameters, those that registered the greatest importance were A, P, L, W, curvature, and burned area, indicating that an analysis which considers morphometric factors, together with soil erosion data affected by water and soil moisture, is an important indicator in the analysis of soil erosion in watersheds.

Forest fires are considered to be an important part of numerous terrestrial ecosystems and vegetation types, being also a significant factor of ecosystem disruption. In this sense, fires play an important role in the structure and function of the ecosystems. Biomes are characterized by a specific type of fire regime, which is a synergy of the climate conditions and the characteristics of the vegetation types dominating each biome. The assessment of burned areas and the identification of the fire regimes can be implemented with freely available low- to high-resolution satellite data as those of Landsat and Sentinel-2. Moreover, the biomes are characterized by the phenology, a useful component for vegetation monitoring, especially when time series of satellite images are used. Both the identification of fire regime by reconstructing the fire history and the monitoring of the post-fire evolution of burned areas were studied with remote sensing methods. Specifically, the present paper is a pilot study implemented in a Mediterranean biome, aimed at establishing the methodological framework to (i) define fire regimes, (ii) characterize the phenological pattern of the vegetation (pre-fire situation) of the fire-affected areas, and (iii) compare the phenology of the recovered fire-affected areas with the corresponding one of the pre-fire situation. At the global level, based on MODIS fire perimeters, we found that fires are occurring at 70% in the tropical and subtropical grasslands, savannas, and shrublands, followed by fires at tropical and subtropical moist broadleaf forests by 7% and by fires at deserts and xeric shrublands by 6.5%.

Forest fire is a common concern in Mediterranean watersheds. Fire-induced canopy mortality may cause the degradation of chemical–physical properties in the soil and influence hydrological processes within and across watersheds. However, the prediction of the pedological and hydrological effect of forest fires with heterogenous severities across entire watersheds remains a difficult task. A large forest fire occurred in 2017 in northern Italy providing the opportunity to test an integrated approach that exploits remote and in-situ data for assessing the impact of forest fires on the hydrological response of semi-natural watersheds. The approach is based on a combination of remotely-sensed information on burned areas and in-situ measurements of soil infiltration in burned areas. Such collected data were used to adapt a rainfall–runoff model over an experimental watershed to produce a comparative evaluation of flood peak and volume of runoff in pre- and post-fire conditions. The model is based on a semi-distributed approach that exploits the Soil Conservation Service Curve Number (SCS-CN) and lag-time methods for the estimation of hydrological losses and runoff propagation, respectively, across the watershed. The effects of fire on hydrological losses were modeled by adjusting the CN values for different fire severities. Direct infiltration measurements were carried out to better understand the effect of fire on soil infiltration capacity. We simulated the hydrological response of the burned watershed following one of the most severe storm events that had hit the area in the last few years. Fire had serious repercussions in regard to the hydrological response, increasing the flood peak and the runoff volume up to 125% and 75%, respectively. Soil infiltration capacity was seriously compromised by fire as well, reducing unsaturated hydraulic conductivity up to 75% compared with pre-fire conditions. These findings can provide insights into the impact of forest fires on the hydrological response of a whole watershed and improve the assessment of surface runoff alterations suffered by a watershed in post-fire conditions.

Mapping burned areas using satellite imagery has become a subject of extensive research over the past decades. The availability of high-resolution satellite data allows burned area maps to be produced with great detail. However, their increasing spatial resolution is usually not matched by a similar increase in the temporal domain. Moreover, high-resolution data can be a computational challenge. Existing methods usually require downloading and processing massive volumes of data in order to produce the resulting maps. In this work we propose a method to make this procedure fast and yet accurate by leveraging the use of a coarse resolution burned area product, the computation capabilities of Google Earth Engine to pre-process and download Sentinel-2 10-m resolution data, and a deep learning model trained to map the multispectral satellite data into the burned area maps. For a 1500 ha fire our method can generate a 10-m resolution map in about 5 min, using a computer with an 8-core processor and 8 GB of RAM. An analysis of six important case studies located in Portugal, southern France and Greece shows the detailed computation time for each process and how the resulting maps compare to the input satellite data as well as to independent reference maps produced by Copernicus Emergency Management System. We also analyze the feature importance of each input band to the final burned area map, giving further insight about the differences among these events.

This paper presents a new method, based on clustering and thresholding, to automatically perform binary change detection in multitemporal spectral indices. The method is denoted as Buffer-From-Cluster Approach (BFCA). To estimate the distributions of changed and unchanged pixels, as needed for the purpose of a reliable thresholding of a spectral index, a clustering algorithm is preliminarily applied to identify image objects possibly corresponding to areas where significant changes occurred. Then, a buffer zone is created around the selected cluster to identify unchanged areas surrounding changed ones. The cluster and the buffer zone are jointly analyzed to estimate the distributions of changed and unchanged pixels and to verify that they can be distinguished from each other. Finally, the results of thresholding and clustering are combined to generate the binary change map. The BFCA has been conceived to map the extent of the areas affected by a natural disaster like wildfire. To validate the proposed method, burned area maps produced by applying the BFCA to spectral indices derived from Sentinel-2 data have been compared to maps produced by the Copernicus Emergency Management Service. For testing the multi-hazard detection capability, the same kind of exercise has been carried out for a flooding test case too. The positive results of the comparison have confirmed the effectiveness of the proposed method.

Wildfires have major ecological, social and economic consequences. Information about the extent of burned areas is essential to assess these consequences and can be derived from remote sensing data. Over the last years, several methods have been developed to segment burned areas with satellite imagery. However, these methods mostly require extensive preprocessing, while deep learning techniques—which have successfully been applied to other segmentation tasks—have yet to be fully explored. In this work, we combine sensor-specific and methodological developments from the past few years and suggest an automatic processing chain, based on deep learning, for burned area segmentation using mono-temporal Sentinel-2 imagery. In particular, we created a new training and validation dataset, which is used to train a convolutional neural network based on a U-Net architecture. We performed several tests on the input data and reached optimal network performance using the spectral bands of the visual, near infrared and shortwave infrared domains. The final segmentation model achieved an overall accuracy of 0.98 and a kappa coefficient of 0.94.

Characterizing biophysical changes in land change areas over large regions with short and noisy multivariate time series and multiple temporal parameters remains a challenging task. Most studies focus on detection rather than the characterization, i.e., the manner by which surface state variables are altered by the process of changes. In this study, a procedure is presented to extract and characterize simultaneous temporal changes in MODIS multivariate times series from three surface state variables the Normalized Difference Vegetation Index (NDVI), land surface temperature (LST) and albedo (ALB). The analysis involves conducting a seasonal trend analysis (STA) to extract three seasonal shape parameters (Amplitude 0, Amplitude 1 and Amplitude 2) and using principal component analysis (PCA) to contrast trends in change and no-change areas. We illustrate the method by characterizing trends in burned and unburned pixels in Alaska over the 2001–2009 time period. Findings show consistent and meaningful extraction of temporal patterns related to fire disturbances. The first principal component (PC1) is characterized by a decrease in mean NDVI (Amplitude 0) with a concurrent increase in albedo (the mean and the annual amplitude) and an increase in LST annual variability (Amplitude 1). These results provide systematic empirical evidence of surface changes associated with one type of land change, fire disturbances, and suggest that STA with PCA may be used to characterize many other types of land transitions over large landscape areas using multivariate Earth observation time series.

The objective of this study was to analyze the spatial and temporal distribution of burned areas in Rondônia State, Brazil during the years 2000 to 2011 and evaluate the burned area maps. A Linear Spectral Mixture Model (LSMM) was applied to MODIS surface reflectance images to originate the burned areas maps, which were validated with TM/Landsat 5 and ETM+/Landsat 7 images and field data acquired in August 2013. The validation presented a correlation ranging from 67% to 96% with an average value of 86%. The lower correlation values are related to the distinct spatial resolutions of the MODIS and TM/ETM+ sensors because small burn scars are not detected in MODIS images and higher spatial correlations are related to the presence of large fires, which are better identified in MODIS, increasing the accuracy of the mapping methodology. In addition, the 12-year burned area maps of Rondônia indicate that fires, as a general pattern, occur in areas that have already been converted to some land use, such as vegetal extraction, large animal livestock areas or diversified permanent crops. Furthermore, during the analyzed period, land use conversion associated with climatic events significantly influenced the occurrence of fire in Rondônia and amplified its impacts.

Fires modify the structure of vegetation communities, the carbon and water cycles, the soil’s chemistry, and affect the climate system. Within this context, this work aimed to understand the distribution patterns of burned areas in Brazil, during the period of 2002 to 2010, taking into consideration each one of the six Brazilian biomes (Amazon, Caatinga, Cerrado, Atlantic Forest, Pampa and Pantanal) and the respective major land cover classes. Data from the MODIS MCD45A1 product (burned area), as well as thermal anomalies (MOD14 and MYD14) and precipitation (TRMM), were analyzed according to the 2002 Brazilian official land cover and land use map (PROBIO). The Brazilian savanna biome, known as Cerrado, presented the largest concentration of burned areas detected by MODIS (73%), followed by the Amazon (14%), Pantanal (6%), Atlantic Forest (4%), Caatinga (3%), and Pampa (0,06%) biomes. Indeed, in the years of 2007 and 2010, 90% and 92% of Brazil’s burned areas were concentrated in the Cerrado and Amazon biomes, respectively. TRMM data indicated that during these two years there was a significant influence of La Niña, causing low rainfall in the Amazon, Cerrado, Caatinga, and Atlantic Forest biomes. Regarding the land cover classes, approximately 81% of the burned areas occurred over remnant vegetation areas. Although no unequivocal correlation can be established between burned areas and new land conversions, the conspicuous concentration of fire scars, particularly in Amazon–Cerrado transition (i.e., the Arc of Deforestation) is certainly not a simple coincidence. Such patterns and trends corroborate the need of improved territorial governance, in addition to the implementation of systematic fire warning and preventive systems.

Mount Kenya is one of Kenya’s ‘water towers’, the headwaters for the country’s major rivers including the Tana River and Ewaso Nyiro River, which provide water and hydroelectric power to the semiarid region. Fires affect water downstream, but are difficult to monitor given limited resources of local land management agencies. Satellite-based remote sensing has the potential to provide long term coverage of large remote areas on Mount Kenya, especially using the free Landsat data archive and moderate resolution imaging spectroradiometer (MODIS) fire products. In this study, we mapped burn scars on Mount Kenya using 30 m Landsat 7 Enhanced Thematic Mapper Plus (ETM+) and Landsat 8 Operational Land Imager (OLI) derived dNBR (change in normalized burn ratio) and MODIS active fire detection and burned area data for fires occurring from 2004 to 2015. We also analyzed topographic position (elevation, slope, aspect) of these fires using an ASTER global digital elevation model (GDEM v2) satellite-derived 30 m digital elevation model (DEM). Results indicate that dNBR images calculated from data acquired about one year apart were able to identify large fires on Mount Kenya that match locations (and timing) of MODIS active fire points and burned areas from the same time period, but we were unable to detect smaller and/or older fires.