Search

Assessing different levels of red gum lerp psyllid (Glycaspis brimblecombei) can influence the hyperspectral reflectance of leaves in different ways due to changes in chlorophyll. In order to classify these levels, the use of machine learning (ML) algorithms can help process the data faster and more accurately. The objectives were: (I) to evaluate the spectral behavior of the G. brimblecombei attack levels; (II) find the most accurate ML algorithm for classifying pest attack levels; (III) find the input configuration that improves performance of the algorithms. Data were collected from a clonal eucalyptus plantation (clone AEC 0144—Eucalyptus urophilla) aged 10.3 months old. Eighty sample evaluations were carried out considering the following severity levels: control (no shells), low infestation (N1), intermediate infestation (N2), and high infestation (N3), for which leaf spectral reflectances were obtained using a spectroradiometer. The spectral range acquired by the equipment was 350 to 2500 nm. After obtaining the wavelengths, they were grouped into representative interval means in 28 bands. Data were submitted to the following ML algorithms: artificial neural networks (ANN), REPTree (DT) and J48 decision trees, random forest (RF), support vector machine (SVM), and conventional logistic regression (LR) analysis. Two input configurations were tested: using only the wavelengths (ALL) and using the spectral bands (SB) to classify the attack levels. The output variable was the severity of G. brimblecombei attack. There were differences in the hyperspectral behavior of the leaves for the different attack levels. The highest attack level shows the greatest distinction and the highest reflectance values. LR and SVM show better accuracy in classifying the severity levels of G. brimblecombei attack. For the correct classification percentage, the RL and SVM algorithms performed better, both with accuracy above 90%. Both algorithms achieved F-score values close to 0.90 and above 0.8 for Kappa. The entire spectral range guaranteed the best accuracy for both algorithms. [ABSTRACT FROM AUTHOR]

Predicting maize yield using spectral information, temperature, and different irrigation management through machine learning algorithms provide information in a fast, accurate, and non-destructive way. The use of multispectral sensor data coupled with irrigation management in maize allows further exploration of water behavior and its relationship with changes in spectral bands presented by the crop. Thus, the objective of this study was to evaluate, by means of multivariate statistics and machine learning techniques, the relationship between irrigation management and spectral bands in predicting maize yields. Field experiments were carried out over two seasons (first and second seasons) in a randomized block design with four treatments (control and three additional irrigation levels) and eighteen sample repetitions. The response variables analyzed were vegetation indices (IVs) and crop yield (GY). Measurement of spectral wavelengths was performed with the Sensefly eBee RTK, with autonomous flight control. The eBee was equipped with the Parrot Sequoia multispectral sensor acquiring reflectance at the wavelengths of green (550 nm ± 40 nm), red (660 nm ± 40 nm), red-edge (735 nm ± 10 nm), and NIR (790 nm ± 40 nm). The blue length (496 nm) was obtained by additional RGB imaging. Data were subjected to Pearson correlations (r) between the evaluated variables represented by a correlation and scatter plot. Subsequently, the canonical analysis was performed to verify the interrelationship between the variables evaluated. Data were also subjected to machine learning (ML) analysis, in which three different input dataset configurations were tested: using only irrigation management (IR), using irrigation management and spectral bands (SB+IR), and using irrigation management, spectral bands, and temperature (IR+SB+Temp). ML models used were: Artificial Neural Network (ANN), M5P Decision Tree (J48), REPTree Decision Tree (REPT), Random Forest (RF), and Support Vector Machine (SVM). A multiple linear regression (LR) was tested as a control model. Our results revealed that Random Forest has higher accuracy in predicting grain yield in maize, especially when associated with the inputs SB+IR and SB+IR+Temp. [ABSTRACT FROM AUTHOR]

Caatinga biome, located in the Brazilian semi-arid region, is the most populous semi-arid region in the world, causing intensification in land degradation and loss of biodiversity over time. The main objective of this paper is to determine and analyze the changes in land cover and use, over time, on the biophysical parameters in the Caatinga biome in the semi-arid region of Brazil using remote sensing. Landsat-8 images were used, along with the Surface Energy Balance Algorithm for Land (SEBAL) in the Google Earth Engine platform, from 2013 to 2019, through spatiotemporal modeling of vegetation indices, i.e., leaf area index (LAI) and vegetation cover (VC). Moreover, land surface temperature (LST) and actual evapotranspiration (ETa) in Petrolina, the semi-arid region of Brazil, was used. The principal component analysis was used to select descriptive variables and multiple regression analysis to predict ETa. The results indicated significant effects of land use and land cover changes on energy balances over time. In 2013, 70.2% of the study area was composed of Caatinga, while the lowest percentages were identified in 2015 (67.8%) and 2017 (68.7%). Rainfall records in 2013 ranged from 270 to 480 mm, with values higher than 410 mm in 46.5% of the study area, concentrated in the northern part of the municipality. On the other hand, in 2017 the lowest annual rainfall values (from 200 to 340 mm) occurred. Low vegetation cover rate was observed by LAI and VC values, with a range of 0 to 25% vegetation cover in 52.3% of the area, which exposes the effects of the dry season on vegetation. The highest LST was mainly found in urban areas and/or exposed soil. In 2013, 40.5% of the region's area had LST between 48.0 and 52.0 °C, raising ETa rates (~4.7 mm day−1). Our model has shown good outcomes in terms of accuracy and concordance (coefficient of determination = 0.98, root mean square error = 0.498, and Lin's concordance correlation coefficient = 0.907). The significant increase in agricultural areas has resulted in the progressive reduction of the Caatinga biome. Therefore, mitigation and sustainable planning is vital to decrease the impacts of anthropic actions. [ABSTRACT FROM AUTHOR]

The aims of this study were: i) to compare no-till areas in two municipalities located in different regions of Brazil, along with the influence on CO2Flux and GPP, and ii) to verify the difference between environmental factors followed by the trends of these variables regarding future scenarios (ARIMA time-series model number). The study was carried out in two areas with different latitudes in the municipalities of Sinop-MT and Passo Fundo-RS, both in Brazil. A time series of 19 years was performed with data acquired by remote sensing from the following satellites: i) Landsat-8 (OLI and TIRS), and ii) TERRA/AQUA (MODIS). The results propound that the spectro-temporal variables are directly influenced by soil management and agricultural practices over the observation time, with a satisfactory correlation in future predictions of the variables for the next ten years, in which presented that the variation of GPP and albedo values for the two study sites would gradually increase until 2028 and the temperature remained constant between the range of its seasonality, and CO2Flux tends to decrease in its seasonality, indicating a higher CO2 absorption. [ABSTRACT FROM AUTHOR]

• Remotely sensed data in establishment of new boundary between biomes. • Land Use and Land Cover Change and the advance of the Cerrado biome on the edge of the Brazilian Amazonia. • Biophysical-atmospheric aspects altered with the anthropic advance and its influence on public environmental policies. The Amazonia is the largest tropical forest in the world, with approximately 5 million km2 occupying about 49.3% of Brazil in the North and Central-West regions and bordering the Cerrado sensu lato (Brazilian savannah). The border area between these two biomes is called the Cerrado-Amazonia transition that extends approximately 6240 km. Currently, these limits, as well as other areas of ecotonal stress, are not clearly defined mainly due to intense Land Use and Land Cover Change (LULCC). Remote sensing is a tool that tries to define and update these limits. In this context, the objective of this study was to use remote sensing tools to determine the transition between the two large Brazilian biomes. The study area comprised the transition between the Cerrado-Amazonia biomes in Brazil and an area composed of a buffer of 300 km was created around the official line of the abovementioned biomes. We used data from some remote sensors to calculate the composition of the previous year's pixel means for the Normalized Difference Vegetation Index (NDVI), Enhanced Vegetation Index (EVI) (-1˜1), CO 2 Flux (μmol m−2 s-1), evaporative efficiency (mm yr-1), and Gross Primary Production (GPP) (g C m−2 d−1) variables. The statistical analyses were performed by softening the pixels of all the variables by means of the Kernel density (KD). Later, the analysis of the main components and the cluster between the biomes and the remote sensing variables were applied. Therefore, we sought to standardize the results and to find an overall mean between the standards to delimit a new area of the Amazonia and Cerrado biomes, called the Transition Area between Biomes (TAB). The variables analyzed by remote sensing were effective for identifying the Amazonia and Cerrado biomes characteristics that highlighted the evaporative efficiency variable in relation to other variables in both biomes. Thus, from the results achieved, a new boundary was determined that delimits the transition of the biomes that is different than the official government boundary. The analyses used in this study constitute important tools to foment the creation of public policies able to curb the anthropic actions on these biomes. [ABSTRACT FROM AUTHOR]

Abstract Brazil is the second largest soybean producer in the world with a planted area in the crop year 2017/18 of 33.347 million hectares, distributed in the Pampa, Atlantic Forest, Cerrado, and Amazon biomes. Through remote sensing techniques we show that the new agricultural frontier of soy is no longer in the Amazon, but in the last continuous areas of Cerrado, present in the region known as MATOPIBA. The soybean production chain has been striving to present to its overseas customers a soy produced in a sustainable way, without the removal of forests. Our data challenge its main program, the Amazonian Soy Moratorium, and we call attention to the conservation need of the MATOPIBA Cerrado, which is not monitored by Soy Moratorium. [ABSTRACT FROM AUTHOR]

Copyright of Canadian Journal of Remote Sensing is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

The Amazon Basin is undergoing extensive environmental degradation as a result of deforestation and the rising occurrence of fires. The degradation caused by fires is exacerbated by the occurrence of anomalously dry periods in the Amazon Basin. The objectives of this study were: (i) to quantify the extent of areas that burned between 2001 and 2019 and relate them to extreme drought events in a 20-year time series; (ii) to identify the proportion of countries comprising the Amazon Basin in which environmental degradation was strongly observed, relating the spatial patterns of fires; and (iii) examine the Amazon Basin carbon balance following the occurrence of fires. To this end, the following variables were evaluated by remote sensing between 2001 and 2019: gross primary production, standardized precipitation index, burned areas, fire foci, and carbon emissions. During the examined period, fires affected 23.78% of the total Amazon Basin. Brazil had the largest affected area (220,087 fire foci, 773,360 km2 burned area, 54.7% of the total burned in the Amazon Basin), followed by Bolivia (102,499 fire foci, 571,250 km2 burned area, 40.4%). Overall, these fires have not only affected forests in agricultural frontier areas (76.91%), but also those in indigenous lands (17.16%) and conservation units (5.93%), which are recognized as biodiversity conservation areas. During the study period, the forest absorbed 1,092,037 Mg of C, but emitted 2908 Tg of C, which is 2.66-fold greater than the C absorbed, thereby compromising the role of the forest in acting as a C sink. Our findings show that environmental degradation caused by fires is related to the occurrence of dry periods in the Amazon Basin. [ABSTRACT FROM AUTHOR]

• Real-time monitoring of boron fertilization in eucalyptus is helpful for guiding precision diagnosis; • The 350–371 nm spectral range can be used for detecting boron-deficient plants; • Adequate boron levels can be identified by using the 426–444, 1811–1910, 1948–2115, and 2124–2208 nm; • The 425–475 nm spectral range can be used to find boron-toxicity plants. Boron (B) is an essential element whose deficiency results in rapid inhibition in the growth of plants, acting on their meristematic growth. Real-time monitoring of B fertilization in eucalyptus is helpful for guiding precision diagnosis and efficient management of plant boron nutrition. This research hypothesizes that different boron levels alter the reflectance of different wavelengths in eucalyptus. In this context, the objective of this study was to identify spectral ranges that can be used to monitor the boron status in eucalyptus plants. The experiment was carried out in a greenhouse, in which the treatments consisted of increasing boron levels in the form of boric acid (17% of B), whose levels varied from deficit to toxicity. Thus, five treatments were established: no boron, 1, 10, 20, and 40 mg/dm3 of boron. The remote sensing data used were bands, heights, and vegetation indices calculated after obtaining the spectral curves in each treatment. Our findings show that it is possible to accurately distinguish the boron levels in eucalyptus using hyper and multispectral bands. The 350–371 nm spectral range can be used for detecting boron-deficient plants. Plants with adequate boron levels can be identified by using the 426–444 nm, 1811–1910 nm, 1948–2115 nm, and 2124–2208 nm spectral ranges. Finally, the 425–475 nm spectral range can be used to find boron-toxicity plants. [ABSTRACT FROM AUTHOR]

Cotton has high production costs compared to other annual crops because large numbers of plant protection product (PPP) applications can be needed to control insect pests, diseases, and growth. The hypothesis underlying this study was that vegetation indices (VIs) could be used to estimate application rates for cotton. Our objectives were to (i) evaluate the relationship between different VIs and the application rates for cotton; (ii) propose a modification to the canopy chlorophyll content index (CCCI); and (iii) to develop a VI based equation that will indicate the ideal application rate needed to maximize deposition in the middle layer of a cotton crop. The experiments were carried out during the crop seasons 2017/18, and 2018/19 in the State of Mato Grosso do Sul, Brazil. A multispectral sensor installed in an unmanned aerial vehicle (UAV) was used to obtain the VIs, and the application rates evaluated were 40, 70, 100, and 130 L ha−1. The spray deposits on cotton leaves were measured using the mass balance analysis method. Our findings revealed that an increase in the VIs led to a rise in the application rate needed to maintain spray deposition on the middle layer of cotton plants. The CCCI is related to the rate variation in the cotton crop. However, our results showed that the proposed modified equation (the simplified modified canopy chlorophyll content index), which is based on the relative deposition, improves the estimation of the application rate that will optimize spray deposition in the middle layer of cotton plants. [ABSTRACT FROM AUTHOR]

Wildfires in the Amazon biome of Mato Grosso cause extensive environmental, economic, and health damages, including biodiversity loss and high greenhouse gas emissions. This study used remote sensing to examine the relationship between fire severity and climatic factors, focusing on dNBR (Differenced Normalized Burn Ratio), precipitation, LST (Land Surface Temperature), SPI (Standardized Precipitation Index), NDVI (Normalized Difference Vegetation Index), and VCI (Vegetation Condition Index), analyzing data from 2001 to 2022. Statistical tests included Shapiro-Wilk, Tukey, regression kriging, Mann-Kendall for trend analysis, Pettitt for change points, and canonical variable tests. Regarding trends, only LST showed a significant trend starting in 2009, with the Northeast mesoregion showing the highest impact on temperature. dNBR correlated positively with NDVI and VCI, and negatively with precipitation and SPI. The northern mesoregion had a positive influence on dNBR and NDVI but negative for precipitation, SPI, and VCI. The southwestern mesoregion associated positively with dNBR and LST but negatively with the other variables. The Northeast and South-Central mesoregions showed positive correlations with most variables except dNBR and NDVI. These findings highlight the northern mesoregion's vulnerability due to its proximity to the central Amazon Forest and agri-cultural activity, indicating increased fire susceptibility with reduced humidity. • The northeast mesoregion of the Mato Grosso Amazon had the greatest impact on temperature. • There is vulnerability in the northern mesoregion due to its proximity to the central rainforest and agricultural activity. • The main positive correlations for dNBR were with the variables NDVI and VCI. [ABSTRACT FROM AUTHOR]

Greenhouse gas (GHG) sources and sinks are an important global concern. Monitoring the spatiotemporal variations of GHG concentrations, particularly carbon dioxide (CO 2), is crucial for identifying potential sources and sinks and moving toward a sustainable future. Therefore, via a time-series of remote data and multispectral images, this study evaluates the CO 2 spatiotemporal dynamics and related factors during 2015–2018 in one of the world's main agricultural frontier areas, the state of Mato Grosso (SMT), Brazil, which is both experiencing continued deforestation and attempting to achieve sustainable food production. In this study, data was obtained from the measurement of column-averaged carbon dioxide (CO 2) dry air mole fraction in the atmosphere, set as X CO2 from Orbiting Carbon Observatory-2 satellite from January 2015 to December 2018. The enhanced vegetation index data were obtained from the Moderate-Resolution Imaging Spectroradiometer (MODIS) sensor, and rainfall data were obtained from the Climate Hazards Group InfraRed Precipitation with Station dataset. From a series of Landsat-8 satellite images, it was possible to distinguish land use and land cover classes and estimate the CO 2 flux in the SMT. The results showed that the temporal variability of CO 2 flux is correlated positively with rainfall, while X CO2 is negatively correlated with rainfall. Regarding spatial variability, we observed that forest areas that were converted to other land uses resulted in higher values that characterize with sources, and that the highest and lowest average concentrations of CO 2 occurred in the dry and rainy months, respectively, for X CO2 , which might be the result of differences in the vertical resolution of the CO 2 column and scale. In contrast, areas with large continuous forest areas tended to have lower values and contribute positively to the carbon balance as sinks, thereby mitigating climate change impacts. Therefore, not only X CO2 but also CO 2 flux are directly related to changes in land use and land cover (LULC) in complex systems that are affected by climatic variables and processes, such as photosynthesis and soil respiration. • X CO2 is inversely related to rainfall, with highest concentration in drier periods. • Human actions in land use and land cover change increase atmospherical CO 2. • Remote sensing to locate and understand the sources and sinks of carbon dioxide. [ABSTRACT FROM AUTHOR]

Land cover mapping over large areas is essential to address a wide spectrum of socio-environmental challenges. For this reason, many global or regional land cover products are regularly released to the scientific community. Yet, the remote sensing community has not fully addressed the challenge to extract useful information from vast volumes of satellite data. Especially, major limitations concern the use of inadequate classification schemes and "black box" methods that may not match with end-users conceptualization of geographic features. In this paper, we introduce a knowledge-driven methodological approach to automatically process Sentinel-2 time series in order to produce pre-classifications that can be adapted by end-users to match their requirements. The approach relies on a conceptual framework inspired from ontologies of scientific observation and geographic information to describe the representation of geographic entities in remote sensing images. The implementation consists in a three-stage classification system including an initial stage, a dichotomous stage and a modular stage. At each stage, the system firstly relies on natural language semantic descriptions of time series of spectral signatures before assigning labels of land cover classes. The implementation was tested on 75 time series of Sentinel-2 images (i.e. 2069 images) in the Southern Brazilian Amazon to map natural vegetation and water bodies as required by a local end-user, i.e. a non-governmental organization. The results confirmed the potential of the method to accurately detect water bodies (F-score = 0.874 for bodies larger than 10 m) and map natural vegetation (max F-score = 0.875), yet emphasizing the spatial heterogeneity of accuracy results. In addition, it proved to be efficient to provide rapid estimates of degraded riparian forests at watershed level (R 2 = 0.871). Finally, we discuss potential improvements both in the system's implementation, e.g. considering additional characteristics, and in the conceptual framework, e.g. moving from pixel- to object-based image analysis and evolving towards a hybrid system combining data- and knowledge-driven approaches. • End-user requirements are rarely taken into account in large scale land cover maps. • We developed a knowlegde-driven approach to automatically map large areas. • We introduced an ontology-based conceptual framework. • We implemented a system to classify Sentinel 2 times series in the Brazilian Amazon. • The system was efficient to map vegetation gradients and small water bodies. [ABSTRACT FROM AUTHOR]

The objectives of this study are: (i) to evaluate the space-temporal variability of fire foci by environmental satellites, CHIRPS and remote sensing products based on applied statistics, and (ii) to identify the relational pattern between the distribution of fire foci and the environmental, meteorological, and socioeconomic variables in the mesoregions of Minas Gerais (MG) - Brazil. This study used a time series of fire foci from 1998 to 2015 via BDQueimadas. The temporal record of fire foci was evaluated by Mann-Kendall (MK), Pettitt (P), Shapiro-Wilk (SW), and Bartlett (B) tests. The spatial distribution by burned area (MCD64A1-MODIS) and the Kernel density - (radius 20 km) were estimated. The environmental variables analyzed were: rainfall (mm) and maximum temperature (°C), besides proxies to vegetation canopy: NDVI, SAVI, and EVI. PCA was applied to explain the interaction between fire foci and demographic, environmental, and geographical variables for MG. The MK test indicated a significant increasing trend in fire foci in MG. The SW and B tests were significant for non-normality and homogeneity of data. The P test pointed to abrupt changes in the 2001 and 2002 cycles (El Niño and La Niña moderated), which contributes to the annual increase and in winter and spring, which is identified by the Kernel density maps. Burned areas highlighted the northern and northwestern regions of MG, Triângulo Mineiro, Jequitinhonha, and South/Southwest MG, in the 3rd quarter (increased 17%) and the 4th quarter (increased 88%). The PCA resulted in three PCs that explained 71.49% of the total variation. The SAVI was the variable that stood out, with 11.12% of the total variation, followed by Belo Horizonte, the most representative in MG. We emphasize that the applied conceptual theoretical model defined here can act in the environmental management of fire risk. However, public policies should follow the technical-scientific guidelines in the mitigation of the resulting socioeconomic - environmental damages. Image 1 • There was a significant increasing trend in fire foci in Minas Gerais. • The North and Northwest regions of MG, Triângulo Mineiro, and Jequitinhonha have more burned areas. • The proposed theoretical-conceptual model can act in environmental fire risk management. [ABSTRACT FROM AUTHOR]