Mudelsee M. Climate Time Series Analysis: Classical Statistical and Bootstrap Methods

Atmospheric and Oceanographic Sciences Library, vol. 42. Springer, Berlin, New York, 2010, 502 pp. - ISBN: 978-90-481-9481-0
Climate is a paradigm of a complex system. Analysing climate data is an exciting challenge. Analysis connects the two other fields where climate scientists work, measurements and models. Climate time series analysis uses statistical methods to learn about the time evolution of climate. The most important word in this book is estimation. We wish to know the truth about the climate evolution but have only a limited amount of data (a time series) influenced by various sources of error (noise). We cannot expect our guess (estimate), based on data, to equal the truth. However, we can determine the typical size of that deviation (error bar). Related concepts are confidence intervals or bias. Error bars help to critically assess estimation results, they prevent us from making overstatements, they guide us on our way to enhance the knowledge about the climate. Estimates without error bars are useless.Acknowledgements.
List of Algorithms.
Part I Fundamental Concepts.Persistence Models.
Bootstrap Confidence Intervals.
Part II Univariate Time Series.
Regression I.
Spectral Analysis.
Extreme Value Time Series.
Part III Bivariate Time Series.
Correlation.
Regression II.
Part IV Outlook.
Future Directions.Subject Index.
Author Index.