Antonio Roberto Formaggio

Relationships of spectral reflectance and color among surface and subsurface horizons of tropical soil profiles

Abstract The reflectance spectra (400–2500 nm) and color attributes of samples from the major horizons of six soil classes with broad distribution in Brazil were analyzed for intrinsic relationships (principal components analysis) as well as for correlations with chemical components. The objective was to detect the spectral changes with depth that are related to the soil chemical properties, thus improving the pedological characterization of the soil profile and vertically extending the links between topsoil properties and remote sensing data. The albedo (first principal component) differentiated the studied soil profiles in all horizons and was inversely related to FeA (total iron), Ti02, and Al203 contents. The spectral variability within each soil profile was associated with changes in the shape (slope) of the spectra (second principal component). Such changes produced ratio values between longer and shorter wavelengths that tended to decrease with depth because of the increasing quantities of clay components, and the interplay between iron and carbon, respectively. Thus, in each profile, the albedo was generally preserved, but the shape of the spectra was not, except for soils with very low albedo. Hematite-rich soils were distinguished from goethite-rich soils by the redness index, which was estimated from the three color parameters: dominant wavelength, purity, and luminance. The results suggested that the use of spectroradiometers to measure reflectance of samples from several horizons can expedite soil surveys, and also optimize laboratory analysis for soil chemical evaluation. Also, they can be important to characterize the relative distribution of some soil constituents in large unvegetated areas.

Dynamic modeling of forest conversion: Simulation of past and future scenarios of rural activities expansion in the fringes of the Xingu National Park, Brazilian Amazon

Abstract The present work is committed to simulate the expansion of agricultural and cattle raising activities within a watershed located in the fringes of the Xingu National Park, Brazilian Amazon. A spatially explicit dynamic model of land cover and land use change was used to provide both past and future scenarios of forest conversion into such rural activities, aiming to identify the role of driving forces of change in the study area. The employed modeling platform – Dinamica EGO – consists in a cellular automata environment that embodies neighborhood-based transition algorithms and spatial feedback approaches in a stochastic multi-step simulation framework. Biophysical variables and legal restrictions drove this simulation model, and statistical validation tests were then conducted for the generated past simulations (from 2000 to 2005), by means of multiple resolution fitting methods. Based on optimal calibration of past simulations, future scenarios were conceived, so as to figure out trends and spatial patterns of forest conversion in the study area for the year 2015. In all simulated scenarios, pasturelands remained nearly stable throughout the analyzed period, while a large expansion in croplands took place. The most optimistic scenario indicates that more than 50% of the natural forest will be replaced by either cropland or pastureland by 2015. This modeling experiment revealed the suitability of the adopted model to simulate processes of forest conversion. It also indicates its possible further applicability in generating simulations of deforestation for areas with expanding rural activities in the Amazon and in tropical forests worldwide.

An investigation of the selection of texture features for crop discrimination using SAR imagery

This article presents a methodology for selecting texture measures to maximize the discrimination of agricultural land use classes in SAR images. The images were acquired during the first flight of the Shuttle Imaging Radar-C (SIR-C) experiment, in April 1994. L (24 cm)- and C (5 cm)-band SAR data at HH (horizontal transmitting and receiving), HV (horizontal transmitting, vertical receiving), and VV (vertical transmitting and receiving) polarizations both in ground range and slant range and in two different passes were analyzed. The kappa statistic was used to identify meaningful texture measures to discriminate seven classes. The results show that the classifications of land use based only on tonal averages produced a kappa coefficient only slightly higher than 0.50. A kappa threshold of 0.90 was reached with the simultaneous inclusion of 15 texture measures for the six images (two bands, three polarizations). It was also found that the inclusion of texture features when only one band and one polarization was used could produce kappa values higher than 0.85.

Landslide inventory using image fusion techniques in Brazil

The present work aims to assess the accuracy of six fusion techniques (Brovey, IHS, HSV, PCA, WTYO and WTVE) in order to compile landslide inventories using orbital images (ETM+ and PAN HRV). The study area is characterized by steep terrain and dense forest in Caraguatatuba, Sao Paulo State, Brazil. In terms of spatial quality, the Wavelet Transform technique provided the best results, presenting correlations above 90%. As for spectral quality, the best results were obtained with the IHS fusion. Based on the results, it may be concluded that the IHS is the best technique for preserving spatial and spectral information from the original images, so as to more clearly identify landslide scars. However, it was still not possible to typify the landslides from remote sensing data. Nonetheless, it is believed that image fusion techniques adequately met expectations in terms of their capacity to identify landslide for the creation of an inventory for the studied area.

Bi-directional reflectance factor of 14 soil classes from Brazil

Abstract The spectral reflectance of soils is required for effective use of remote sensing products. The absence of studies concerned with spectral reflectance of the soils from the tropical region in the 400 to 2500 nm spectral range is the main motivation of this research. The objective of this study was to present spectral reflectance data from different tropical soil types. This spectral characterization was done through measurements of the bi-directional reflectance factor of 111 selected soil samples, grouped in 14 tropical soil classes, taken from 53 sites (Sao Paulo State, Brazil). The measurements were made with a spectroradiometer operating in the 400 to 2500 nm region of the electromagnetic spectrum. Each soil sample is associated to a set of physical and chemical analyses data, with part of these published in descriptive reports of soil surveys.

Self-Guided Segmentation and Classification of Multi-Temporal Landsat 8 Images for Crop Type Mapping in Southeastern Brazil

Only well-chosen segmentation parameters ensure optimum results of object-based image analysis (OBIA). Manually defining suitable parameter sets can be a time-consuming approach, not necessarily leading to optimum results; the subjectivity of the manual approach is also obvious. For this reason, in supervised segmentation as proposed by Stefanski et al. (2013) one integrates the segmentation and classification tasks. The segmentation is optimized directly with respect to the subsequent classification. In this contribution, we build on this work and developed a fully autonomous workflow for supervised object-based classification, combining image segmentation and random forest (RF) classification. Starting from a fixed set of randomly selected and manually interpreted training samples, suitable segmentation parameters are automatically identified. A sub-tropical study site located in Sao Paulo State (Brazil) was used to evaluate the proposed approach. Two multi-temporal Landsat 8 image mosaics were used as input (from August 2013 and January 2014) together with training samples from field visits and VHR (RapidEye) photo-interpretation. Using four test sites of 15 × 15 km2 with manually interpreted crops as independent validation samples, we demonstrate that the approach leads to robust classification results. On these samples (pixel wise, n ≈ 1 million) an overall accuracy (OA) of 80% could be reached while classifying five classes: sugarcane, soybean, cassava, peanut and others. We found that the overall accuracy obtained from the four test sites was only marginally lower compared to the out-of-bag OA obtained from the training samples. Amongst the five classes, sugarcane and soybean were classified best, while cassava and peanut were often misclassified due to similarity in the spatio-temporal feature space and high within-class variabilities. Interestingly, misclassified pixels were in most cases correctly identified through the RF classification margin, which is produced as a by-product to the classification map.

Hidden Markov Models for crop recognition in remote sensing image sequences

This work proposes a Hidden Markov Model (HMM) based technique to classify agricultural crops. The method uses HMM to relate the varying spectral response along the crop cycle with plant phenology, for different crop classes, and recognizes different agricultural crops by analyzing their spectral profiles over a sequence of images. The method assigns each image segment to the crop class whose corresponding HMM delivers the highest probability of emitting the observed sequence of spectral values. Experimental analysis was conducted upon a set of 12 co-registered and radiometrically corrected LANDSAT images of region in southeast Brazil, of approximately 124.100ha, acquired between 2002 and 2004. Reference data was provided by visual classification, validated through extensive field work. The HMM-based method achieved 93% average class accuracy in the identification of the correct crop, being, respectively, 10% and 26% superior to multi-date and single-date alternative approaches applied to the same data set.

Estimating soybean crop areas using spectral-temporal surfaces derived from MODIS images in Mato Grosso, Brazil

The objective of this work was to evaluate the application of the spectral-temporal response surface (STRS) classification method on Moderate Resolution Imaging Spectroradiometer (MODIS, 250 m) sensor images in order to estimate soybean areas in Mato Grosso state, Brazil. The classification was carried out using the maximum likelihood algorithm (MLA) adapted to the STRS method. Thirty segments of 30x30 km were chosen along the main agricultural regions of Mato Grosso state, using data from the summer season of 2005/2006 (from October to March), and were mapped based on fieldwork data, TM/Landsat-5 and CCD/CBERS-2 images. Five thematic classes were considered: Soybean, Forest, Cerrado, Pasture and Bare Soil. The classification by the STRS method was done over an area intersected with a subset of 30x30-km segments. In regions with soybean predominance, STRS classification overestimated in 21.31% of the reference values. In regions where soybean fields were less prevalent, the classifier overestimated 132.37% in the acreage of the reference. The overall classification accuracy was 80%. MODIS sensor images and the STRS algorithm showed to be promising for the classification of soybean areas in regions with the predominance of large farms. However, the results for fragmented areas and smaller farms were less efficient, overestimating soybean areas.

Predicting forest fire in the Brazilian Amazon using MODIS imagery and artificial neural networks.

The presented work describes a methodology that employs artificial neural networks (ANN) and multi-temporal imagery from the MODIS/Terra-Aqua sensors to detect areas of high risk of forest fire in the Brazilian Amazon. The hypothesis of this work is that due to characteristic land use and land cover change dynamics in the Amazon forest, forest areas likely to be burned can be separated from other land targets. A study case was carried out in three municipalities located in northern Mato Grosso State, Brazilian Amazon. Feedforward ANNs, with different architectures, were trained with a backpropagation algorithm, taking as inputs the NDVI values calculated from MODIS imagery acquired during five different periods preceding the 2005 fire season. Selected samples were extracted from areas where forest fires were detected in 2005 and from other non-burned forest and agricultural areas. These samples were used to train, validate and test the ANN. The results achieved a mean squared error of 0.07. In addition, the model was simulated for an entire municipality and its results were compared with hotspots detected by the MODIS sensor during the year. A histogram analysis showed that the spatial distribution of the areas with fire risk were consistent with the fire events observed from June to December 2005. The ANN model allowed a fast and relatively precise method to predict forest fire events in the studied area. Hence, it offers an excellent alternative for supporting forest fire prevention policies, and in assisting the assessment of burned areas, reducing the uncertainty involved in currently used methods.

Directional effects on NDVI and LAI retrievals from MODIS: A case study in Brazil with soybean

Abstract The Moderate Resolution Imaging Spectroradiometer (MODIS) is largely used to estimate Leaf Area Index (LAI) using radiative transfer modeling (the “main” algorithm). When this algorithm fails for a pixel, which frequently occurs over Brazilian soybean areas, an empirical model (the “backup” algorithm) based on the relationship between the Normalized Difference Vegetation Index (NDVI) and LAI is utilized. The objective of this study is to evaluate directional effects on NDVI and subsequent LAI estimates using global (biome 3) and local empirical models, as a function of the soybean development in two growing seasons (2004–2005 and 2005–2006). The local model was derived from the pixels that had LAI values retrieved from the main algorithm. In order to keep the reproductive stage for a given cultivar as a constant factor while varying the viewing geometry, pairs of MODIS images acquired in close dates from opposite directions (backscattering and forward scattering) were selected. Linear regression relationships between the NDVI values calculated from these two directions were evaluated for different view angles (0–25°; 25–45°; 45–60°) and development stages ( 90 days after planting). Impacts on LAI retrievals were analyzed. Results showed higher reflectance values in backscattering direction due to the predominance of sunlit soybean canopy components towards the sensor and higher NDVI values in forward scattering direction due to stronger shadow effects in the red waveband. NDVI differences between the two directions were statistically significant for view angles larger than 25°. The main algorithm for LAI estimation failed in the two growing seasons with gradual crop development. As a result, up to 94% of the pixels had LAI values calculated from the backup algorithm at the peak of canopy closure. Most of the pixels selected to compose the 8-day MODIS LAI product came from the forward scattering view because it displayed larger LAI values than the backscattering. Directional effects on the subsequent LAI retrievals were stronger at the peak of the soybean development (NDVI values between 0.70 and 0.85). When the global empirical model was used, LAI differences up to 3.2 for consecutive days and opposite viewing directions were observed. Such differences were reduced to values up to 1.5 with the local model. Because of the predominance of LAI retrievals from the MODIS backup algorithm during the Brazilian soybean development, care is necessary if one considers using these data in agronomic growing/yield models.