Page 42 - EER-2-1

Explora: Environment

and Resource

ORIGINAL RESEARCH ARTICLE
New method for statistical analysis of climate

time series

Eric Zeltz*
Independent Scholar, La Motte en Champsaur, Hautes-Alpes, France

Abstract

After publishing four articles utilizing a new method for the statistical study of climate
time series, we found it useful to provide a detailed review of the method itself, which
is the primary objective of this work. Unlike the methods most commonly used by
scientists analyzing such data, this new method does not seek to identify trends for
explorative forecasts. Instead, it enables the detection of precise signals indicating
interactions with other climate entities, thereby enhancing our understanding
of the underlying phenomena. As illustrated through three example articles, the
mechanisms uncovered using this method can be integrated into a mathematical
model. The simulations thus obtained are more deterministic than stochastic – a
significant advantage for producing high-quality forecasts in the context of global
warming. Even if this was the sole application of the method, it would be sufficient
to demonstrate its value. However, as a final example detailed in this work shows,
reconsidering the original series using different periods (e.g., month, quarter,
*Corresponding author:
Eric Zeltz semester, year) can further refine our understanding of the mechanisms at play.
(ericzeltz@wanadoo.fr) We conclude this work by exploring the potential applicability of this method for
Citation: Zeltz E. New method for analyzing non-climatic temporal data series.
statistical analysis of climate time
series. Explora Environ Resour.
2025;2(1):6109. Keywords: Climate time series; Markov chains; Signals; Statistical analysis method
doi: 10.36922/eer.6109
Received: November 17, 2024
1st revised: December 21, 2024 1. Introduction
2nd revised: January 14, 2025
Time series theory has enabled great advances in sciences as well as in econometrics,
1
Accepted: January 16, 2025 information theory, demography, and astronomy. 2-5
Published online : February 6, The application of this theory facilitates the isolation of trends as well as the
2025
identification of value stability and variations within the analyzed series. Based on
Copyright: © 2025 Author(s). this approach, it is possible to generate robust future or past projections, which often
This is an Open-Access article
distributed under the terms of the demonstrate greater reliability compared to those derived from even highly sophisticated
Creative Commons Attribution structural models.
License, permitting distribution,
and reproduction in any medium, A major challenge in climatology is to obtain, from data collected in the form of time
provided the original work is
properly cited. series, projections on the future of the parameters in question, accompanied by error
bars, without which these projections would not be of real use. 6
Publisher’s Note: AccScience
Publishing remains neutral with The classic approach begins with a climate equation that projects the climatic variable
regard to jurisdictional claims in X(t) at time step t as the sum of a “trend” component X (t) and a “noise” component
published maps and institutional trend
affiliations. X noise (t), which reflects, for example, natural variability.

Volume 2 Issue 1 (2025) 1 doi: 10.36922/eer.6109