Detrended fluctuation analysis as a statistical tool to monitor the climate

TL;DR

This paper demonstrates that detrended fluctuation analysis can effectively identify and differentiate climate regions in the western US by analyzing long-term temperature fluctuation trends and their power law exponents.

Contribution

It introduces a novel application of detrended fluctuation analysis to climate monitoring, linking power law exponents with regional climate distinctions.

Findings

Different geographical regions show distinct power law exponents.

Power law exponents correlate with temperature fluctuation variability.

Long-term temperature trends can distinguish climate types.

Abstract

Detrended fluctuation analysis is used to investigate power law relationship between the monthly averages of the maximum daily temperatures for different locations in the western US. On the map created by the power law exponents, we can distinguish different geographical regions with different power law exponents. When the power law exponents obtained from the detrended fluctuation analysis are plotted versus the standard deviation of the temperature fluctuations, we observe different data points belonging to the different climates, hence indicating that by observing the long-time trends in the fluctuations of temperature we can distinguish between different climates.