Massive data production is a critical aspect of experimental sciences. We have experienced exponential growth in data availability in the past decade, and it has not been different for geoscience. Examples of geoscientific data include any physical observable related to the energy industry, mining, monitoring hazardous areas (e.g., effects of mining in populated areas), etc. Nowadays, with the relative facility and lowering the cost to acquire data – even in continuous mode – the data processing to exploit their value is a challenge. Also, it requires expertise in data maintenance and processing, data analysis, and the design of experiments of target domains for which data will provide insight and knowledge.

Traditional analysis in the area using these computing and data services revolves around the creation of physics-based models. These are developed and executed by geoscience professionals with a deep understanding of geology, geophysics, engineering, reservoir dynamics, production technology, and economics, among other things. Data managers support these geoscience professionals to ensure high-quality data that powers the decision-making process. In contexts with a high data density, significant actions are now being invested in advanced analytical techniques such as machine learning to augment decisions traditionally made exclusively by geoscientists and engineers.

Many scientists and companies believe that they can generate fresh insight, reduce decision cycle times, and steal a march on their competition by automating the search for patterns and relationships in their data. Therefore, geophysics and data science, including algorithms, mathematical models, and computing, must converge for developing experiments for obtaining insight and foresight about the observations contained in data collections. Furthermore, experiments represent best practices for addressing problems and questions on geophysics that must be treated as data and knowledge to be shared and reused by scientists and practitioners.

OBJECTIVE

The ADAGEO Summit and Thematic School aim at promoting scientific discussion and practical actions and projects that can address geosciences problems through data science solutions. Thereby, the activities and interaction in this event will build a scientific and practitioners’ transdisciplinary community the will develop a novel way of doing geosciences.

Topics

• Applying statistical methods to investigate and unveil new geophysical data patterns, answer open problems, or lead to further research questions.
• Building predictive models to describe better or approximate geophysical phenomena, increasing our knowledge about our planet.
• Parallelizing algorithms for processing geophysical data, allowing for the processing of extensive data sets in reasonable times.

Experiments will provide the ground to devise new data curation approaches to define new techniques for processing heterogeneous data. In addition, statistical knowledge is essential for extracting information from the massive amount of data we will process. New methods and models will be crucial to model data and make conclusions in a timely fashion.

• Bruno M. Carvalho, UFRN, Brazil
• Umberto Costa, UFRN, Brazil
• Martin Musicante, UFRNS, Brazil
• Aderson Nascimento, UFRN, Brazil
• Marcus Alexandre Nunes, UFRN, Brazil
• Genoveva Vargas-Solar, CNRS, France

Data Science Challenges Couches

• Flavio Lemos de Santana, UFRN, Brazil
• José Fonsêca, UFRN, Brazil
• Javier A. Espinosa Oviedo, University of Lyon, France
• Karen Maria Leopoldino Oliveira, UFC, Brazil
• Placido Antonio Souza Neto, IFRN, Brazil
• Samuel Oliveira, UFRN, Brazil
• Waldson Patricio, UFRN, Brazil