This article presents a new method for monitoring forest cover in the state of Rondônia, in the Brazilian Amazon. The proposed method applies the Linear Spectral Mixing Model (LSMM) to Landsat datasets (MSS, TM and OLI) to derive annual vegetation, soil, and shade fraction images for the period 1980 – 2020. These fraction images have the advantages of reducing the volume of data to be analyzed and highlighting the target characteristics. Then, we applied a threshold method to classify forest, non-forest, hydrography, and deforestation areas. The proposed method showed to be consistent and flexible allowing to change the threshold values according to the fraction images to obtain the results with high accuracy. The results obtained by the proposed method can be easily checked over the RGB image mosaic. This kind of information is very important for environmental and climate change studies and for supporting government conservation efforts.

This article presents a land use and land cover (LULC) classification map based on Random Forest (RF) classifier algorithm in the São Paulo State (Brazil), using Landsat-8 OLI data. The method consists in using time series images from January to December of 2020 based on the spectral and temporal characteristics of the LULC classes. We performed the classification class by class considering: water, urban area, forest, agriculture, forest plantation and pasture. Then, we pre-processed the selected images based on the spectral characteristics of the targets to highlight each LULC class. After that, the classification was performed using RF for each class individually and then we composed the final map with all LULC classes. The results showed a global accuracy of 89.10%, kappa value of 0.8692, producer accuracies greater than 79.80% and user accuracies greater than 76.82% for the classes mapped. Therefore, the method is consistent allowing to minimize the classification errors facilitating the pos-classification edition of individual classes mapped.

This article presents a method to map the extent of forest plantation in an area located in the São Paulo State (Brazil). The proposed method applies the Linear Spectral Mixing Model (LSMM) to Landsat Thematic Mapper (TM) datasets to derive annually vegetation, soil and shade fraction images for local analysis. We used 30 m annual mosaics of TM images during the 1985 to 1995 time period. These fraction images have the advantage to reduce the volume of data to be analyzed highlighting the target characteristics. Then, we generated only one mosaic for each fraction images for TM dataset computing de maximum value through this period, facilitating the classification of areas occupied by forest plantation. The proposed method allowed to classify two forest plantation classes: Eucalypt and Pine. In addition, it allowed to monitor the phenological stages of Eucalypt according to its growth cycle. The results are very important for planning and management by the commercial companies and can contribute to develop an automatic method to map forest plantation areas in a regional and global scales.

This article presents a land use and land cover (LULC) classification map using Random Forest algorithm in the São Paulo State (Brazil), and an assessment of burned areas using two products (MCD64A1 and MapBiomas Fire). The method uses Landsat Operational Land Imager (OLI) time series images from January to December of 2020. We performed the classification class by class considering: water, urban area, forest formation, sugarcane, agriculture, forest plantation and pasture. For each class, we used different spectral bands and image fraction according to the best response for the class. For 2020, the top three areas mapped in São Paulo State were pasture (40.49%), sugarcane (24.74%) and forest formation (20.60%). Comparing the two burned area products, MCD64A1 mapped more burned areas as it uses MODIS images combined with 1 km active fire observations with higher temporal resolution than MapBiomas Fire. About 60% of the burned areas mapped in 2020 occurred in the sugarcane class. The results show the importance of land use and land cover classification for better understanding fire-prone classes given the spatial distribution. It turns as an environmental tool for environmental strategies of planning and monitoring burned area assessment over regional scales.

Unmanned aerial vehicles (UAVs) have been advancing in precision and cost-benefit for remote sensing studies, including height and biomass estimations. This article presents a preliminary experiment to explore UAV photogrammetry to estimate canopy height in savanna and grassland phytophysiognomies in the Brazilian Cerrado biome. For this purpose, it was generated dense cloud points to obtain digital terrain and surface models used to calculate the canopy height. The spatial distribution of canopy height ranged from 0 to 4 meters, in which the most values were under 50 cm, commonly found in grasslands and observed in vegetation regrowth of post-fire event in savanna areas.

Fire dynamics in the Brazilian Savannas (Cerrado) is related to climatic conditions and management interventions by human activities. Thus, the fire occurrence in conservation units (UCs) may be different when compared with their buffer zones. Our results, obtained by burned area analysis, demonstrate that buffer zones have the most significant variation in the burned area over the years when compared to the burned area inside the UC, such as the Jalapão State Park. In contrast, when the buffer zone presents agricultural activities, as occurs in the Chapada dos Veadeiros and Chapada das Mesas National Parks, or urban occupation, such as the Brasilia National Park, the proportions of burned area are lower than found inside the UCs, according to data obtained from the MODIS MCD64A1 product from 2001 to 2018 years.

This article presents a method to map the extent of annual land-use and land-cover (LULC) in Mato Grosso State, located in the Brazilian Legal Amazon. The proposed method applies the Linear Spectral Mixing Model (LSMM) to VIIRS NPP dataset to derive monthly vegetation, soil and shade fraction images for regional analysis. We used 500 m monthly image mosaics for VIIRS in 2015 year. These fraction images have the advantage to reduce the volume of data to be analyzed highlighting the target characteristics. Then we generated only one mosaic for each fraction images for VIIRS dataset computing de maximum value through the year, facilitating the classification of LULC classes. The proposed method allowed to classify three LULC classes: forest, cropland and non-forest (Savannah and pasture) areas. In addition, it allowed to map burned areas occurred during the study period. The results are very important for planning and management by the government and non-governmental organizations.

In this work we present a procedure to analyze the forest degradation by fire in the Brazilian Amazon using the Landsat-8 Operational Land Imager (OLI) time series, taking advantage of the resources of the Google Earth Engine platform. The study area is the municipality of Porto dos Gaúchos located in the state of Mato Grosso, in the "arc of deforestation" of the Brazilian Legal Amazon. We used OLI images acquired between January 1st, 2017 and the last available image from 2018. We generated fraction images of soil, vegetation and shade using the Linear Spectral Mixing Model to highlight burned forest that we visited in the field in September 2018. Our analysis showed that forest degradation by fire can be detected using time series and the Google Earth Engine platform.

The objective of this paper is to present a method to assess the extent of annual land use/land cover in Brazil, South America. The proposed method applies the Linear Spectral Mixing Model (LSMM) to PROBA-V datasets to derive vegetation, soil and shade fraction images for global and regional analysis. We used 1 km composites of 10 days (S10-TOC - 10-daily global composites, Top-Of-Canopy) for the South America and 100 m composites of 5 days (S5-TOC - 5-daily global composites, Top-Of-Canopy) for the Mato Grosso State, Brazilian Amazon. Then we built the 1km and 100m composites corresponding to the three endmembers with the highest fraction values during the year 2015. In that manner we could detect and map the areas occupied by main crops in Brazil, during the 2015 year, using the vegetation fraction composites. Also, PROBA-V images were acquired in the dry season, on 21 June, 26 July and 11 August 2015 to show the potentiality of these images to assess the land cover changes due to deforestation and forest degradation by fire. The agricultural areas mapped using 1km dataset were compared with 100m results for the Mato Grosso State showing a difference of 12% (58,834 km2and 66,490 km2, respectively). The results are very important for the government and nongovernmental organizations for planning and management of the tropical environment.