Integrating Data Science and Earth Science: Challenges and Solutions (SpringerBriefs in Earth System Sciences)

✍ Scribed by Laurens M. Bouwer (editor), Doris Dransch (editor), Roland Ruhnke (editor), Diana Rechid (editor), Stephan Frickenhaus (editor), Jens Greinert (editor)

• Publisher: Springer
• Year: 2022
• Tongue: English
• Leaves: 158
• Category: Library

✦ Synopsis

This open access book presents the results of three years collaboration between earth scientists and data scientist, in developing and applying data science methods for scientific discovery. The book will be highly beneficial for other researchers at senior and graduate level, interested in applying visual data exploration, computational approaches and scientifc workflows.

✦ Table of Contents

Foreword by Johann-Christoph Freytag
Foreword by Hans Pfeiffenberger
Acknowledgments
Contents
1 Data Science and Earth System Science
1.1 Introduction
1.2 Data Science
1.3 Earth System Science
1.4 Challenges
1.5 Digital Earth Culture
References
2 The Digital Earth Project: Focus and Agenda
2.1 History of the Project
2.2 Focus of Digital Earth
2.2.1 Data Analysis and Exploration
2.2.2 Data Collection and Monitoring
2.2.3 Collaborative Interdisciplinary Working
3 Data Analysis and Exploration with Visual Approaches
3.1 Challenges
3.2 The Data Analytics Software Framework (DASF) Providing Linkable Visualization Components
3.2.1 Introduction
3.2.2 Visualization Concept
3.2.3 Technical Implementation
3.2.4 Application
3.3 The Digital Earth Viewer
3.3.1 Introduction
3.3.2 Visualization Concept
3.3.3 Applications
3.4 Spatially Immersive Visualization of Complex Seafloor Terrain
3.4.1 Introduction
3.4.2 Technical Implementation
3.4.3 Application
3.5 Assessment of the Three Visualization Approaches and Techniques
References
4 Data Analysis and Exploration with Computational Approaches
4.1 Introduction and Challenge
4.2 Object Recognition Using Machine Learning
4.2.1 Deep Learning Support for Identifying Uncharted Levees in Germany
4.2.2 Machine Learning Support for Automated Munition Detection in the Seabed
4.3 Approximating Complex Processes with Machine Learning
4.3.1 Estimation of Methane and Ethane Concentrations in the Atmosphere Over Europe by Means of a Neural Network
4.3.2 Fusing Highly Heterogeneous Data to Facilitate Supervised Machine Learning in the Context of Health and Climate Research
4.4 Point-To-Space Extrapolation
4.4.1 Estimation of Missing Methane Emissions from Offshore Platforms by a Data Analysis of the Emission Database for Global Atmospheric Research (EDGAR)
4.4.2 Point-to-Space Methods
4.5 Anomaly and Event Detection
4.5.1 Computational Methods for Investigating the Impacts of the Elbe Flood 2013 on the German Bight
4.6 Conclusions
References
5 Data Analysis and Exploration with Scientific Workflows
5.1 Challenges and Needs
5.2 Scientific Workflows
5.2.1 The Concept of Scientific Workflows
5.2.2 Scientific Workflows in Digital Earth
5.2.3 Digital Implementation of Scientific Workflows with the Component-Based Data Analytics Software Framework (DASF)
5.3 The Digital Earth Flood Event Explorer—A Showcase for Data Analysis and Exploration with Scientific Workflows
5.3.1 The Showcase Setting
5.3.2 Developing and Implementing Scientific Workflows for the Flood Event Explorer
5.3.3 The Workflows of the Flood Event Explorer
5.4 Assessment of the Scientific Workflow Concept
References
6 The Digital Earth Smart Monitoring Concept and Tools
6.1 Challenges
6.2 SMART Monitoring Concept
6.2.1 An Expanded SMART Monitoring Concept
6.2.2 Pre-Conditions for SMART Monitoring
6.2.3 Future Tasks to Further Increase Smart Monitoring Efforts
6.3 SMART Monitoring approaches and tools
6.3.1 Hard-and Software Tools for a Modern Communication between Sensor and Control to Enhance Traditional Monitoring Efforts
6.3.2 SMART DataFlow
6.3.3 SMART MetaData: Without Trustworthy Descriptions, Data can be Un-FAIR
6.3.4 SMART Sampling Approaches
References
7 Interdisciplinary Collaboration
7.1 Challenges
7.2 Material and Methods
7.3 Results and Discussion
7.4 Conclusions and Outlook
References
8 Evaluating the Success of the Digital Earth Project
8.1 Objective
8.2 Approach for Evaluation in the Digital Earth Project
8.3 Evaluation Criteria
8.3.1 Capacities for Doing Data Science
8.3.2 Scientific and Project Goals
8.3.3 Usability of Results
8.4 Results from the Questionnaires
8.5 Conclusions
Appendix: Survey Questions
References
9 Lessons Learned in the Digital Earth Project
9.1 Introduction
9.2 Lesson 1: Interdisciplinary Collaboration
9.3 Lesson 2: Thinking Out of the Box
9.4 Lesson 3: Thinking in Workflows
9.5 Lesson 4: Sustainable Implementation of Scientific Software, Data Infrastructure and Policies