The influence of data gaps and outliers on resilience indicators

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Liu, Teng, Morr, Andreas, Bathiany, Sebastian, Blaschke, Lana L., Qian, Zhen, Diao, Chan, Smith, Taylor, Boers, Niklas
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909033910239232
author Liu, Teng
Morr, Andreas
Bathiany, Sebastian
Blaschke, Lana L.
Qian, Zhen
Diao, Chan
Smith, Taylor
Boers, Niklas
author_facet Liu, Teng
Morr, Andreas
Bathiany, Sebastian
Blaschke, Lana L.
Qian, Zhen
Diao, Chan
Smith, Taylor
Boers, Niklas
contents The resilience, or stability, of major Earth system components is increasingly threatened by anthropogenic pressures, demanding reliable early warning signals for abrupt and irreversible regime shifts. Widely used data-driven resilience indicators based on variance and autocorrelation detect `critical slowing down', a signature of decreasing stability. However, the interpretation of these indicators is hampered by poorly understood interdependencies and their susceptibility to common data issues such as missing values and outliers. Here, we establish a rigorous mathematical analysis of the statistical dependency between variance- and autocorrelation-based resilience indicators, revealing that their agreement is fundamentally driven by the time series' initial data point. Using synthetic and empirical data, we demonstrate that missing values substantially weaken indicator agreement, while outliers introduce systematic biases that lead to overestimation of resilience based on temporal autocorrelation. Our results provide a necessary and rigorous foundation for preprocessing strategies and accuracy assessments across the growing number of disciplines that use real-world data to infer changes in system resilience.
format Preprint
id arxiv_https___arxiv_org_abs_2505_19034
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle The influence of data gaps and outliers on resilience indicators
Liu, Teng
Morr, Andreas
Bathiany, Sebastian
Blaschke, Lana L.
Qian, Zhen
Diao, Chan
Smith, Taylor
Boers, Niklas
Adaptation and Self-Organizing Systems
Chaotic Dynamics
Data Analysis, Statistics and Probability
Geophysics
The resilience, or stability, of major Earth system components is increasingly threatened by anthropogenic pressures, demanding reliable early warning signals for abrupt and irreversible regime shifts. Widely used data-driven resilience indicators based on variance and autocorrelation detect `critical slowing down', a signature of decreasing stability. However, the interpretation of these indicators is hampered by poorly understood interdependencies and their susceptibility to common data issues such as missing values and outliers. Here, we establish a rigorous mathematical analysis of the statistical dependency between variance- and autocorrelation-based resilience indicators, revealing that their agreement is fundamentally driven by the time series' initial data point. Using synthetic and empirical data, we demonstrate that missing values substantially weaken indicator agreement, while outliers introduce systematic biases that lead to overestimation of resilience based on temporal autocorrelation. Our results provide a necessary and rigorous foundation for preprocessing strategies and accuracy assessments across the growing number of disciplines that use real-world data to infer changes in system resilience.
title The influence of data gaps and outliers on resilience indicators
topic Adaptation and Self-Organizing Systems
Chaotic Dynamics
Data Analysis, Statistics and Probability
Geophysics
url https://arxiv.org/abs/2505.19034