A5055768874- Meteorological Phenomena and Simulations
- Hydrological Forecasting Using AI
- Wind and Air Flow Studies
- Climate variability and models
- Atmospheric aerosols and clouds
- Atmospheric and Environmental Gas Dynamics
- Precipitation Measurement and Analysis
- Neural Networks and Applications
- Robotics and Sensor-Based Localization
- Calibration and Measurement Techniques
- Structural Health Monitoring Techniques
- Radiative Heat Transfer Studies
- Fluid Dynamics and Turbulent Flows
- Optical Imaging and Spectroscopy Techniques
- Advanced Multi-Objective Optimization Algorithms
- Metaheuristic Optimization Algorithms Research
- Numerical methods in inverse problems
- Energy Load and Power Forecasting
- Solar Radiation and Photovoltaics
- Geophysics and Gravity Measurements
- Robotic Path Planning Algorithms
- Statistical Mechanics and Entropy
- Fault Detection and Control Systems
- Geophysical and Geoelectrical Methods
- Marine and coastal ecosystems
National Institute for Space Research
2016-2025
Instituto Tecnológico de Aeronáutica
2008-2023
Universidade Federal do Oeste do Pará
2016-2023
Laboratório Nacional de Computação Científica
2023
ORCID
2022
Universidade Federal de Santa Maria
2004-2022
Linköping University
2020
National University of Trujillo
2019
University of Jos
2000-2018
Universidade de Caxias do Sul
2018
Tropical forests concentrate the largest diversity of species on planet and play a key role in maintaining environmental processes. Due to importance those forests, there is growing interest mapping their components getting information at an individual tree level conduct reliable satellite-based forest inventory for biomass distribution qualification. Individual crown could be manually gathered from high resolution satellite images; however, achieve this task large-scale, algorithm identify...
Abstract. We present a new version of the Brazilian developments on Regional Atmospheric Modeling System (BRAMS), in which different previous versions for weather, chemistry, and carbon cycle were unified single integrated modeling system software. This also has set state-of-the-art physical parameterizations greater computational parallel memory usage efficiency. The description main model features includes several examples illustrating quality transport scheme scalars, radiative fluxes...
Experimentally characterizing evapotranspiration (ET) in different biomes around the world is an issue of interest for areas science. ET natural Brazilian Pampa biome has still not been assessed. In this study, actual (ETact) obtained from eddy covariance measurements over two sites was analyzed. The objective to evaluate energy partition and seasonal variability biome. Results showed that latent heat flux dominant component available both autumn–winter (AW) spring–summer (SS) periods....
Many natural disasters in South America are linked to meteorological phenomena. Therefore, forecasting and monitoring climatic events fundamental issues for society various sectors of the economy. In last decades, machine learning models have been developed tackle different society, but there is still a gap applications applied physics. Here, evaluated precipitation prediction over America. Currently, numerical weather unable precisely reproduce patterns due many factors such as lack...
This paper proposes an image matching system using aerial images, captured in flight time, and geo-referenced images to estimate the Unmanned Aerial Vehicle (UAV) position a situation of Global Navigation Satellite System (GNSS) failure. The is based on edge detection posterior automatic registration these edge-images (position estimation UAV). process performed by Artificial Neural Network (ANN), with optimal architecture. A comparison Sobel Canny extraction filters also provided. obtained...
Machine learning has experienced great success in many applications. Precipitation is a hard meteorological variable to predict, but it strong impact on society. Here, machine-learning technique—a formulation of gradient-boosted trees—is applied climate seasonal precipitation prediction over South America. The Optuna framework, based Bayesian optimization, was employed determine the optimal hyperparameters for gradient-boosting scheme. A comparison between forecasting among numerical...
This work presents a comparison of three different techniques to solve the inverse heat conduction problem involving estimation unknown initial condition for one-dimensional slab, whose solution is obtained through minimization known functional form. The following are employed problem: conjugate gradient method with adjoint equation, regularized using quasi-Newton method, and via genetic algorithm (GA) method. For first one, general form compute (considering time space domains) presented, GA...
A machine learning (ML)-based methodology for predicting frosts was applied to the southern and southeastern regions of Brazil, as well other countries including Uruguay, Paraguay, northern Argentina, Bolivia. The model (using TensorFlow (TF)) compared frost index (IG from Portuguese: Índice de Geada) developed by National Institute Space Research (INPE, Brazil). IG is estimated using meteorological variables a regional weather numerical (RWNM). After calculating two indices ML RWNM, voting...
Abstract. Accurate modeling of aerosol optical properties is critical to simulate radiative effects. However, uncertainties regarding the simulation intensive are still significant. Therefore, use observations constrain in models has been indicated as an option. Also, explicit computations too costly for operational models, which make observational-based prescriptions a convenient solution. We developed prescription driven by machine-learning techniques that can be applied models. The...
In this work, two new entropic regularization techniques are introduced. They represent a generalization of the standard MaxEnt method, and allow for greater flexibility introducing any prior information about expected structure true physical model, or its derivatives, into inversion procedure. The first technique is based on minimization entropy vector first-differences unknown parameters. Adopting terminology, it known as minimum first-order method (MinEnt-1). To illustrate essential...
This article describes a methodology for using neural networks in an inverse heat conduction problem. Three network (NN) models are used to determine the initial temperature profile on slab with adiabatic boundary condition, given transient distribution at time. is ill-posed one-dimensional parabolic problem, where condition has be estimated. addressed problem: feedforward backpropagation, radial basis functions (RBF), and cascade correlation. The input NN obtained from set of probes equally...