Updating with new information

Applied geophysics has been developing very rapidly and the applications on civil engineering are growing significantly in the world and also in Brazil. Geophysical methods that have been used on engineering projects, that require accuracy and reliability, are mainly: seismic, resistivity and ground penetrating radar (GPR) methods.

Self potential and some well logging geophysical tools, can also be applied in some designs.

Case studies will be presented. The importance of performing an interpretation of data based on direct information (boreholes, trenches and sampling), as well as, the correlation between mechanics parameters of soils and rocks and physical properties measured by geophysical methods will also be discussed.

The Machine Learning Algorithms is a valuable tool to automatically recognize patterns among high-dimensional data to mitigate bias and to speed up interpretations, especially in regions where geophysical, geological, mineral resources, and remote-sensing data are widely available. Therefore, these techniques represent an efficient method for producing mineral prospectivity maps, where are suitable to recognize characteristics of known deposits.

This training course will provide an introduction to Data-Driven Mineral Potential Modelling, presenting some basic concepts, focused to understand how the main supervised machine-learning algorithms understand the mineral deposits features, and how we can make predictions on an area. The course will also have a hands-on part, using free open-source software, combined with practical tips!

Note: this course will be presented in Brazilian Portuguese.

Target audience:
Geoscience researchers with little or intermediate knowledge in Mineral Potential Modeling and Machine Learning Methods

This training course will provide an introduction to modern laboratory-based techniques applied to the regional geophysical data of South America. This course is aimed at geologists wishing to improve their skill base in modern integrated Structural Geophysics mapping techniques. Many regions in the world now are covered at a high- resolution by airborne geophysical data sets, including magnetic, electromagnetic, digital terrain models and radiometric surveys. When combined with multi-spectral satellite data, and of course the available geological observations, these geophysical data provide key constraints on our geological interpretation, in particular in ancient terrains.

This training course is organised and presented by Mark Jessell (UWA) and will consist of a series of overview lectures, as well as group discussions using South America regional geophysical datasets.

Environmental magnetism involves analysis of the magnetic properties of environmental materials to address questions of interest in Earth, ocean, and climate science. In this course, we will go through the basics of mineral magnetism.

How and why do minerals have magnetic properties? What controls the magnetic properties of minerals? What can the magnetic properties of minerals tell us about about their source, transportation, deposition, or post-depositional alteration and associated environmental processes? Environmental magnetism has diverse applications and has been used to study monsoons, eolian dust, ocean bottom currents, ice-rafting, biomagnetism, pollution, soil development, catchment erosion, sediment provenance, cave flooding, riverine processes, diagenesis, and many more topics. This course is suitable for anyone interested in learning how mineral magnetism can be useful in paleoclimate studies.

A seismic amplitude volume can provide rich and varied information about the subsurface, allowing the mapping of geological structures, and even predicting lateral changes in depositional facies in the subsurface. A clear understanding of the reservoir stratigraphy and the facies distribution can validate the presence of a potential reservoir.

During the stratigraphic interpretation workflow, we can face several challenges to map all the wealth of information that seismic data contains. With the solutions that Emerson E&P Software provides, interpreters are able to validate interpretations, better understand stratigraphic sequences and provide information about the distribution of properties and their relationship to the deposition history.

In this short course, we will discuss different approaches to perform stratigraphic interpretation and show techniques for seismic attributes combination, attributes extraction on horizons, interpretation on flattened volumes, calculation of facies maps/volumes and a volumetric interpretation based on the Wheeler Diagram.

Full Waveform Inversion (FWI) has become a key part of the seismic imaging workflow in the oil industry since a decade. It also spread in many other fields as earthquake seismology, medical imaging, civil engineering, helioseismology,… FWI aims to estimate the subsurface parameters governing seismic wave propagation by fitting each wiggle contained in the seismic data volume. This full-waveform data fitting procedure confers on FWI a high resolution power (of the order of the propagated wavelength) and the ability to reconstruct several parameter classes governing seismic wave propagation. However, the oscillatory nature of seismic waves makes also FWI highly nonlinear due to the cycle skipping pathology. Moreover, the fact that the subsurface is mostly illuminated by surface acquisition device of limited spread can make FWI rather ill-posed. Finally, the computational burden induced by the complete solution of the wave equation for dense multi-source acquisition is a last issue to be considered in particular when FWI is pushed at high frequencies and more realistic visco-elastic physics is involved. All these issues required an in-depth understanding of the theory underlying FWI for a relevant practice of this technology.

The first aim of this lecture is to review the fundamentals of FWI, which include the formulation of the problem as a constrained optimization problem, its classical formulation on a reduced-search space after variable elimination, its solution in the framework of local optimization methods, the physical interpretation of the gradient of the misfit function and the underlying approximation and the role of the Hessian, and the efficient computation of the gradient with the adjoint-state method. From this theoretical introduction, the resolution power of FWI and its relationship with the acquisition design will be illustrated and the condition triggering cycle skipping will be explained. To mitigate the footprint of cycle skipping, heuristic data continuation strategies and preconditioning will be reviewed and illustrated with real data case studies. Finally, I will review the challenge of multi-parameter FWI, discuss the role of the subsurface parametrization and the Hessian, the issue of attenuation reconstruction and illustrate the concepts with synthetic and real data case studies.

In a second part of the lecture, I will address two more advance topics making use of the augmented Lagrangian method and the alternating-direction method of multipliers (ADMM). The first one is related to a versatile implementation of regularization in FWI including nonsmooth and hybrid regularizers. The second one will discuss a reformulation of FWI with an extended search space to mitigate the cycle skipping issue. The review of this extended formulation will cast a new light on the principles governing FWI.

A integração das superfícies de referência para altitudes e profundidades é um requisito essencial para a modelagem dos processos morfodinâmicos costeiros, para a avaliação correta dos riscos de inundações costeiras causadas por eventos meteorológicos extremos e elevação do nível do mar global associada às mudanças climáticas, e para o desenvolvimento das respectivas estratégias de adaptação e mitigação. O desenvolvimento de um modelo topo-batimétrico digital unificado de alta resolução requer a integração de dados topográficos e batimétricos, eliminando descontinuidades devidas a diferenças entre as respectivas superfícies de referência, geralmente associadas aos níveis médio e mínimo de maré, respectivamente. No Brasil, ainda não é possível integrar com precisão as superfícies de referência das elevações dadas pela cartografia terrestre e as profundidades apresentadas nas cartas náuticas, devido à falta de informações necessárias.

Para resolver este problema, as redes fundamentais de altitudes e gravidade do Sistema Geodésico Brasileiro (SGB) ao longo da costa devem ser recuperadas, densificadas e complementadas com outros tipos de medições (nível do mar, aerogravimetria, altimetria oceânica, batimetria rasa de alta resolução etc.), permitindo o refinamento de modelos hidrodinâmicos, gravimétricos e topográficos. Só depois de atingir estes três objetivos será possível fazer uma transformação rigorosa entre os vários níveis de referência exigidos na zona costeira (datum vertical geodésico, níveis de redução náuticos, atuais níveis médios- mínimos-máximos etc.) e determinar uma “linha de inundação” ideal como interseção entre o nível de água máximo projetado e a superfície topográfica. Trata-se, portanto, de uma tarefa interinstitucional e multidisciplinar, sendo absolutamente essencial o nivelamento dos conhecimentos entre as diversas áreas de conhecimento.

Este minicurso abordará os conceitos geodésicos envolvidos na integração altimetria- batimetria (ALT-BAT) na zona costeira brasileira: {1} evolução cronológica e resumo das características das realizações da componente vertical do SGB (ajustamentos altimétricos); {2} breve discussão da aplicação dos GNSS ao posicionamento vertical; e {3} modelos para conversão das altitudes geométricas (GNSS) em altitudes compatíveis com o datum vertical do SGB, com ênfase no novo modelo hgeoHNOR2020.

Migration of seismic data primarily serves the purpose of producing an image of the subsurface that is used by geologists for interpretation and characterization. Migration is a one-step process that maps data from the acquisition domain (source, receiver and time coordinates) to a spatial domain (spatial volume). Least-squares imaging is an iterative procedure that tries to unravel the acquisition geometry and changes in illumination effects from the image.

The course first introduces basic principles of least-squares imaging and how this method relates to the seismic processing and velocity model building workflows, followed by a review of objectives. A second part focuses on describing some of the most common approaches of least-squares imaging from recent literature, and a discussion of benefits as well as pitfalls. The third section focuses on least-squares imaging for migrated gathers and reviews the topic of seismic amplitudes after migration which is important for AVO/AVA studies and seismic inversion investigations. The course concludes with an introduction to other special related topics including the relationship between least-squares imaging and full waveform-inversion, and least-squares imaging of data with multiple energy. The course emphasizes the understanding of fundamental concepts, good practices and pitfalls when applying the technology with real data examples, and encourages class participation.

In the most general context, the Inverse Problem is at the very core of the scientific method; in this regard, when mathematical models are developed to be applied to Science and Engineering problems, by the use of partial or differential equations, solutions will emerge in terms of a system of linear or nonlinear algebraic equations. It follows that in the realm of Science, these solutions or working hypotheses need to be tested and eventually validated by the use of collected data; on the other hand, they could be developed and used for an application in an Engineering problem. In any case, and within the framework of Inverse Theory, we go to the field and/or to the lab to collect data to further establish the value of certain unknown variables m (model parameters) from the measurement values of the so-named variables in d (measured data). In our presentation, m and d will represent vectors that belongs to finite dimensional linear vector spaces, connected by the matrix A; this represents the model for a system of linear equations, d=Am. For the non-linear model, the solution will be addressed by iterating over a linear scheme.

The linear Inverse Problem will be presented with the introduction of the concept of the Generalized Inverse, which applies to any rectangular matrix and is given in terms of its Singular Value Decomposition; also, it will be presented its relationship with methods such as the Least Square and Maximum Likelihood methods. Theory is planned to be presented and discussed with the aid of some science and engineering practical problems, such as: Image forming from a Tomographic Inversion, Normal Mode Inversion, Earthquake location, Pandemic Models, etc.

O curso irá apresentar como os métodos geofísicos marinhos de batimetria, sonar de varredura lateral e sísmica de alta-resolução são utilizadas no mapeamento geológico e geotécnico do fundo para os diferentes tipos de projetos de engenharia submarina. Serão apresentados alguns estudos de caso mostrando como essas ferramentas são essenciais para fornecer informações para a avaliação dos riscos associados à instalação de todos os tipos de estruturas no fundo do mar. 

The course will present how the marine geophysical methods of bathymetry, side scan sonar and high-resolution seismic are used in geological and geotechnical mapping of the seafloor for different types of subsea engineering projects. Some case studies will be presented showing how these tools are essential to provide information for the risk assessment associated with the installation of all kinds of subsea structures on the seabed.

Palaeomagnetism is unique in the Earth Sciences in providing observational data, derived directly from a deep Earth process, back through geological time. The process in question is the geodynamo which emerges as a consequence of conductive outer core liquid convecting in a rapidly rotating shell. The outer core has been subject to external forcing throughout its history: by the nucleation and growth of the underlying inner core, and by variable core-mantle heat flow conditions dictated by the slow convection of the overlying solid rocky mantle. The study of palaeomagnetic records therefore has the potential to reveal information about, not only the ancient geomagnetic field and the geodynamo process responsible for it, but also the secular evolution of the Earth’s deep interior.

This mini-course, requiring no previous experience in palaeomagnetism, will outline the different types of palaeomagnetic measurements that are used to describe the ancient geomagnetic field and also present and discuss the records of palaeomagnetic variations that they comprise. Recent attempts to reproduce statistical properties of the palaeomagnetic field using numerical simulations of the geodynamo will also be discussed along with the potential implications for the study of Earth’s deep interior and for other research fields that use palaeomagnetism.

Side Scan Sonar (SSS) is one the most amazing geophysical technologies for underwater investigation. From SSS images it is possible to see details of sea bottom, as geological and geotechnical features, nautical debris and so on, that make this geophysical technique a very powerful tool for underwater projects like pipelines, cables, wind farms, bridges, tunnels, waterways and search.   

The main target of this training course is to present the fundamentals of side scan sonar technique applied for shallow water investigation (rivers, lakes, water reservoirs and sea - harbors and inner continental shelf). We are going to talk about data acquisition, processing and data interpretation.

For data processing & interpretation we are planning to introduce the attendees to SonarWiz, one the most popular softwares for Side Scan Sonar data acquisition, processing and interpretation.

Importante remarks:
The training course will run in Portuguese.
Some SSS data will be shared with the attendees.
The attendees will be able to run SonarWiz on their own computer.

Time-lapse (4D) seismic has become an integral part of reservoir management in several oil companies. Successful published case studies have demonstrated that 4D seismic can provide useful information to improve reservoir development strategy and maximize recovery factor.

Recently, the feasibility of applying S4D to the Brazilian Pre-Salt oil fields was confirmed. So, it is expected to use Time-lapse seismic intensively from now on due to its potential economic impact, considering the significant volumes of hydrocarbons and high drilling costs.

This course is intended to provide an overview of time-lapse seismic method to create the base for dealing with 4D projects. The course is structured to be didactic and to provide information in a clear, objective and illustrated way to facilitate the understanding of the subject.

Learning objectives
• Introduction – the value of 4D
• Seismic basic concepts
• Seismic acquisition for 4D studies
• Repeatability indicators (NRMS and predictability)
• Seismic processing for 4D studies (an overview)
• Petrophysical basis for 4D: saturation and pressure effects on seismic data
• Feasibility studies to determine whether 4D will work
• Time-Lapse seismic interpretation and case studies
• Time-shifts, S4D-Geomechanics
• S4D applied in carbonate reservoirs (Brazilian Pre-Salt included)



Who should attend?
This course is intended to appeal to a wide audience. It should be suitable not only to geophysicists, but also to geologists, reservoir engineers, petrophysicists and others wanting to know what time-lapse seismic is and how it can help reservoir management. There are no prerequisite degrees or courses required to gain insight from this course.

Since 2006, with the discovery of the pre-salt supergiant oilfields in Santos basin, several seismic technologies have been implemented in the exploration and production development phases. In this course, lessons learned and best practices, improvement points arising from the implementation of these technologies will be discussed, highlighting aspects of acquisition, processing and interpretation of seismic data. Topics of seismic characterization and monitoring of pre-salt carbonate reservoirs in the Campos and Santos basins will also be addressed. The main seismic technologies implemented in pre-salt reservoirs will be presented with the aim to solving geological challenges and improve the quality, imaging, resolution, and detectability in 4D applications. Finally, some perspectives will be presented, as well as new technologies and their respective opportunities, associated with applications in Brazilian pre-salt reservoirs.

Traditionally, most seismic interpreters perform a qualitative seismic interpretation. Seismic reflectors are mapped if based on the geometric structure, whether in the domain of time or depth, however, emphasis has not always been placed on the physical understanding of variations in seismic amplitude. However, in recent decades, some seismic interpreters have emphasized to more quantitative seismic interpretation techniques, as they can validate hydrocarbon anomalies and provide additional information during potential assessment and reservoir characterization. Among some important quantitative techniques we have the AVO technique, which provides important information for oil exploration such as the detection of hydrocarbons, lithological identification and analysis of fluid parameters. Through Halliburton Landmark's software suite (DSG and Seismic Analysis) it is possible to perform the AVO flow in pre-stack and post-stack data.

The course will be taught in Portuguese

Multiphysics imaging combines different data types to improve subsurface exploration and monitoring. This is because individual geophysical methods on their own provide non-unique models of subsurface property and fluid-type present, but integrating them together with geological models maximizes accuracy, minimizes uncertainty in a SHARED EARTH model, and leads to a consistent prognosis for the sought resource or environmental system. This course will introduce the attendees to the state-of-the art practice of multiphysics integration in energy, mining and environmental industries and equip them with the basic tools to drive applications in other fields of geoscience. The topics covered include:
1. Principles of Multiphysics imaging (understanding data from different measurement platforms, their homogenization and integration),
2. Near-surface Multiphysics imaging for environmental, mining and renewable energy investigations,
3. Deep Multiphysics imaging for play-based (basin-scale to prospect-scale) hydrocarbon exploration in different geological environments, and
4. Multiphysics technologies for monitoring hydrocarbon production, C0 2 storage and remediation of contaminated land.

At the end of this course, attendees will: Understand what is meant by the term Multiphysics imaging or integration, understand the physico-chemical linkages used for resolving imaging challenges faced in subsurface investigations for resource (hydrocarbon, geothermal, groundwater, mineral) and environmental (contaminated land, mine collapse) evaluation, understand how/why geophysical models can be integrated to reduce uncertainty, understand the value rather than the cost of Multiphysics information, understand why/how Multiphysics methods are used for reservoir surveillance (4D production monitoring, C0 2 storage/sequestration) and monitoring of remediation in contaminated land studies.

A common challenge regarding velocity modeling is the preservation of the structural and stratigraphic seismic interpretation. The use of a sealed geologic model that mimics the subsurface, capable of including any type of complex structures, such as salt bags, reserve faults, intrusive bodies, and other multi-z interpretation, is key when interpolating interval velocity within stratigraphic layers. Thus, the generation of an accurate geologic model for time-depth conversion workflows can significantly impact the resulting seismic images.

This short course will propose techniques to integrate the structural framework with useful velocity modeling tools, allowing the velocity property to follow the stratigraphic layers. Also, the training will provide strategies for the velocity treatment and analysis within stratigraphic compartments. The proposed velocity modeling workflows could be applied to build more constrained and reliable images of the subsurface.

Course Program:
The course is composed of 4 blocks addressing the following topics: Basic Python Language and Jupyter Notebook introduction. Using Obspy to access waveforms and events data and metadata. Using Obspy for pre-processing steps of your work. Some selected Python examples.

Course Requisites:
Our course will depend on the use of the Google Colab Notebook infrastructure (colab.research.google.com). Google Colab is a programming environment provided by Google to its registered users. This would minimize the need for additional software to be installed on the attendants’ computers. Our view is that the course should be balanced between 25% theoretical and 75% practical. We would like to take advantage of learning by doing concepts.
Also, students are expected to have some degree of experience with programming concepts in languages like Matlab, Python, Fortran or even C. Must understand the concept of variables and functions.

Abstract:
The natural source Magnetotelluric (MT) is becoming an attractive electromagnetic method to explore deep tectonic settings as well as very shallow targets in mineral and geothermal explorations. Thanks to recent advances in instrumentation and software developments, the MT is a standard method in many geophysical applications. In this course we will give an introduction into the method and show examples from around the world to explore structures from the uppermost mantle to near surface mineral and geothermal applications. The course will cover:

• Theoretical background of the method: Introduction to EM methods
• Data processing: From time series to transfer functions
• Analysing the transfer functions for dimensionality
• The inversion theory in EM.
• 3D inversion of MT data using our code; ModEM
• Practical remarks on the 3D inversion of MT data
• Examples from different places on the globe and various scales

Rock physics and seismic inversion are essential tools for reservoir characterization. Rock physics improves the knowledge of how variations in reservoir properties affect the seismic response whereas seismic inversion is a sturdily integrating tool to obtain the 3D volume of elastic properties. In the scope of rock physics, this short course discusses the effect of the depositional environment on the elastic property and, additionally, it examines the estimation of rock properties from elastic properties.

Seismic inversion brings the benefits from dealing with layer property while extending the seismic resolution in compare with seismic amplitude. Statistical foundations underpin the integration of the concepts of rock physics and seismic inversion, thereby revealing the uncertainties under seismic analysis. Besides, the outcomes of seismic-driven statistical estimations are confronting by laboratory tests and dynamic data to mitigate the risk in seismic interpretation. This lecture aims at practical situations in reservoir characterization. To this end, it presents a plethora of study cases to clarify the applications of the quantitative seismic interpretation. These study cases encompass turbidite and pre-salt reservoirs, chiefly in the Brazilian basins.

Rock Physics is the link between seismic amplitudes and geological properties and processes. The over- simplistic use of AVO can lead to overconfidence and failure, and understanding rock physics is key to the successful use of AVO when de-risking prospects. A key difference between carbonate and siliciclastic rock physics modeling and feasibility for prospect de-risking is in the relatively greater stiffness implied by the carbonate mineralogy, the more advanced diagenetic cementation, and the more complex pore microstructure.

This course presents effective considerations for interpreting seismic AVO, how rock physics is used to make those interpretations, and workflow best practices. The course begins with review of systematic and objective Quality Availability Consistency (QAC) process for log data before, during, and after petrophysical interpretation and seismic data before, during and after pre-inversion conditioning to ensure reliable inputs for geophysical applications.

The course covers rock physics model calibration in carbonates followed by perturbation for scenario testing, log repair, forward modeling for AVO inversion feasibility and post inversion reservoir characterization. Various examples representing pre-salt carbonate reservoirs will then be analyzed and discussed.

O curso será composto de duas partes. Na primeira parte, serão disponibilizados materiais em vídeo curtos (até 30 minutos) para estudo individual e acontecerá antes da data devidamente agendada para as atividades do curso. Os vídeos serão acompanhados de indicações de leitura e um questionário de fixação (é recomendada uma dedicação de até uma hora por dia para acompanhar este material). Nos dois dias indicados, iremos estar discutindo os tópicos apresentados e desenvolvendo exercícios práticos pela pela plataforma Submachine.

Durante todo o período ficaremos disponíveis para tirar dúvidas e para eventuais discussões, tanto via plataforma Google/Hangouts quanto por e-mail. Ao participar deste curso, você também estará incentivando a expansão da plataforma Submachine para incorporar modelos de propriedades físicas sob a América do Sul.

Geological modeling is a fundamental step in the characterization of petroleum reservoirs, having as main objective the structural framework with the relations of the depositional system facies and their respective petrophysical properties. The assessment of geological scenarios in an integrated environment within a single platform allows the use of inputs generated from exploration to development, making geological modeling processes more complex as new information is added . The most important step in this process is faciological modeling, seeking to represent the heterogeneities of formations using geophysics as a quantitative and/or qualitative support . Given the need to represent these variables in a geocellular model, geostatistical techniques followed by the distribution of petrophysical properties are used for a more accurate estimate of the volume of oil in place . In subsequent stages of reservoir evaluation, the models produced here will be used to define field development strategies, where the integration of static and dynamic data contributes to a better accuracy in production forecasting.

This short course will cover all the steps mentioned above for the construction of a geological model from an integrated geology-geophysics assessment.

Note: this course will be presented in Brazilian Portuguese.

En los últimos años, se ha establecido como una fuente importante de información la exploración de las técnicas desarrolladas usando el cálculo de la sísmica pasiva. Los métodos de ruido sísmico ambiental han contribuido de manera eficiente para la construcción de modelos de velocidades y el cálculo del efecto de sitio en zonas urbanas para el análisis del Vs30.

En este curso, comprenderemos el estado del arte que surge a partir de un conjunto de pequeñas vibraciones de ondas Rayleigh que se originan en la superficie de la Tierra. Estas vibraciones de origen natural son registradas en sismógrafos que miden los movimientos horizontales y verticales del suelo. Estos procesos surgen como consecuencia de la instrumentación, los avances de la computación y la teoría de la física estadística.

La técnica H/V (también conocida como técnica de Nakamura) ha sido eficiente para calcular los periodos dominantes de vibración del suelo en zonas urbanas y ha sido referente en muchos lugares del mundo. El método de SPAC (Spatial Autocorrelation Method) propuesto por Aki (1957) registra el ruido sísmico ambiental en un conjunto de estaciones sísmicas y calcula modelos de velocidades de ondas de corte a profundidad usando el cálculo de las correlaciones cruzadas de las ondas superficiales.

Estos avances han impulsado grandes aplicaciones en la Ingeniería Sísmica y Geotecnia, implementando aplicaciones innovadoras para establecer las propiedades mecánicas del subsuelo y su geometría. Permitiendo calcular escenarios de la respuesta sísmica considerando la geología superficial.

O Ojetivo do curso é mostrar os fluxos integrados da Emerson disponíveis no Integrated Canvas(EPOS) e como esses fluxos podem entregar alta performance e principalmente diminuir o tempo gasto na caracterização de áreas complexas ou não. Ainda neste mesmo curso mostraremos nossas 3 soluções de Inteligencia Artificial aplicada a sísmica e como ela pode nos ajudar a detectar e predizer fácies em qualquer ambiente geológico.

Data are becoming more readily available for a variety of applications from subsurface characterization to hazard assessment, and the modern geoscientist will need to be well-versed in computational approaches to take advantage of ‘big data’. While many data analysis and machine learning courses focus on theory, in this course we aim to provide background and exercises to address real-world challenges and applications in the geosciences.
You will become familiar with solving geoscience problems using the Python programming language and Jupyter notebooks. We will cover techniques in data analysis, modeling, and machine learning for subsurface characterization applicable to resource assessment, with a significant focus on integrating ML approaches with geological domain knowledge.