Computational Methods for Time-Series Analyses in Earth Sciences

by Silvio José Gumiere Ph.D., Hossein Bonakdari

Overview

Computational Methods for Time-Series Analyses in Earth Sciences bridges the gap between theoretical knowledge and practical application, offering a deep dive into the utilization of R programming for managing, analyzing, and forecasting time-series data within the realm of Earth sciences. It systematically unfolds the layers of data manipulation, graphical representation, and sampling to prepare the reader for complex analyses and predictive modeling from the basics of signal processing to the nuances of machine learning. It presents cutting-edge techniques, such as neural networks, kernel-based methods, and evolutionary algorithms, specifically tailored to tackle challenges, and provides practical case studies to aid readers with utilizing the techniques covered.Computational Methods for Time-Series Analyses in Earth Sciences is a valuable resource for scientists, researchers, and students delving into the intricacies of Earth's environmental patterns and cycles through the lens of computational analysis and guides readers through various computational approaches to deciphering spatial and temporal data. - Focuses on the use of R for time-series analysis and the application of these methods directly to Earth and environmental datasets - Integrates Machine Learning techniques, enabling readers to explore advanced computational methods for forecasting and modeling - Includes case studies with real-world applications, providing readers with examples on how to translate computational skills into tangible outcomes

Product Details

ISBN-13:	9780443336324
Publisher:	Elsevier Science
Publication date:	06/20/2025
Sold by:	Barnes & Noble
Format:	eBook
Pages:	420
File size:	100 MB
Note:	This product may take a few minutes to download.

About the Author

Prof. Silvio José Gumiere has been Professor at the Department of Soils and Agri-Food Engineering, Laval University, Canada, since 2011. He is an expert on the application of R-based numerical, statistical, and geostatistical methods, such as time series analyses, image and signal processing, erosion modeling, spatial hydrology, and spatial interpolation methods. His research has been published in international journals and conferences. He is an editor for several journals on hydrological modeling and machine learning techniques for solving applied science problems in hydrology, soil sciences, soil hydrology, and environmental journals.Dr. Hossein Bonakdari is a distinguished professor in the Department of Civil Engineering at the University of Ottawa, specializing in mathematical modeling and artificial intelligence (AI). A leading expert in AI-driven data analysis, he has pioneered advanced algorithms for real-time forecasting and big data interpretation, significantly improving the understanding and management of environmental systems. Dr. Bonakdari has authored four books, published over 320 peer-reviewed journal articles, contributed to more than 20 book chapters, and delivered over 100 presentations at national and international conferences. As a respected editorial board member of several leading journals, he continues to shape research in his field. His groundbreaking contributions have earned him global recognition, ranking him among the top 2% of the world's scientists from 2019 to 2024.

Table of Contents

Section 1: Theory and Computational Methods1. Introduction to R: Data manipulation, graphics, and sampling2. Time series analysis for earth sciences with R3. Signal processing with R for earth sciences.4. Spatial Analyses with R for earth sciences5. Deterministic modelling with R for earth sciences6. Machine learning with R for earth sciencesSection 2: Case of Studies and Applications7. Predicting Sandy Soils' Hydraulic Properties and Drainage Capacities with Neural Networks8. Prognostication of Real-Time Hourly Precipitation using Kernel-based Techniques9. Integrating Upstream Runoff and Local Rainfall for Real-Time Flood Prediction10. Pre-diagnosis of Flooding Using Real-Time Monitoring of Climate Parameters11. Comparing Local vs. External Data Analysis for Forecasting12. Evolutionary Kernel Extreme Learning Machine for Real-Time Forecasting13. A Stochastic AI Method for Predicting Climatic Variables' Spatio-Temporal Changes Under Future Climates – Data Preparation and Preprocessing14. A Novel AI Stochastic Approach for Predicting Spatio-Temporal Variables and Changes Under Future Climate Conditions: Google Earth Engine's Benefits and Challenges; An Intro to SOILPARAM APP15. A Novel AI Stochastic Method for Predicting Changes in Space and Time: Linear Modeling16. A Novel AI Stochastic Method for Predicting Changes: Nonlinear Modeling17. A Combination of Satellite Observations and Machine Learning Technique for Terrestrial Anomaly Estimation