forecasting
Article
Visual Analytics for Climate Change Detection in
Meteorological Time-Series
Milena Vuckovic * and Johanna Schmidt
Citation: Vuckovic, M.; Schmidt, J.
Visual Analytics for Climate Change
Detection in Meteorological
Time-Series. Forecasting 2021, 3,
276–289. https://doi.org/10.3390/
forecast3020018
Academic Editor: Umberto Triacca
Received: 19 March 2021
Accepted: 16 April 2021
Published: 19 April 2021
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Copyright: © 2021 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
VRVis Zentrum für Virtual Reality und Visualisierung Forschungs-GmbH, 1220 Vienna, Austria;
johanna.schmidt@vrvis.at
* Correspondence: milena.vuckovic@vrvis.at
Abstract: The importance of high-resolution meteorological time-series data for detection of trans-
formative changes in the climate system is unparalleled. These data sequences allow for a com-
prehensive study of natural and forced evolution of warming and cooling tendencies, recognition
of distinct structural changes, and periodic behaviors, among other things. Such inquiries call for
applications of cutting-edge analytical tools with powerful computational capabilities. In this regard,
we documented the application potential of visual analytics (VA) for climate change detection in
meteorological time-series data. We focused our study on long- and short-term past-to-current
meteorological data of three Central European cities (i.e., Vienna, Munich, and Zürich), delivered
in different temporal intervals (i.e., monthly, hourly). Our aim was not only to identify the related
transformative changes, but also to assert the degree of climate change signal that can be derived
given the varying granularity of the underlying data. As such, coarse data granularity mostly offered
insights on general trends and distributions, whereby a finer granularity provided insights on the
frequency of occurrence, respective duration, and positioning of certain events in time. However,
by harnessing the power of VA, one could easily overcome these limitations and go beyond the
basic observations.
Keywords: climate change; meteorological time-series; global warming; visual analytics; visual
computing
1. Introduction
1.1. Background
The unprecedented global increase in the frequency and magnitude of extreme weather
events and related consequences (e.g., heat waves, flooding and drought, severe storms,
wildfires) is being recognized as one of the most pressing environmental issues and a
worldwide health and lifestyle concern. The synthesis reports published by the Intergov-
ernmental Panel on Climate Change (IPCC) confirmed that these observed transformative
changes in climate system are closely tied to the anthropogenic processes and related
elevated emissions of greenhouse gases (GHG) in the Earth’s atmosphere [1,2]. It is a
well-documented fact that GHG such as water vapor, carbon dioxide (CO
2
), methane
(CH
4
), nitrous oxide, which occur naturally in the atmosphere, along with the synthetic
fluorinated gases, which originate from a variety of industrial processes, have the tendency
to absorb, store and reradiate long-wave radiation emitted from Earth’s surface back to
Earth’s surface [3,4]. The effect is generally known as the ‘greenhouse effect’ and has a
significant impact on energy budget of the Earth system, resulting in global atmospheric
warming and chaotic weather patterns worldwide [3]. The global character of this phe-
nomenon is mainly driven by the fact that Earth’s atmosphere intermixes globally, meaning
that this phenomenon is of no geographical or spatial specificity. However, the degree to
which this drives site-specific environmental issues, such as the air, water and soil pollution,
alongside the occurrence of landslides, fluvial flooding, wildfires in forest landscapes, to
Forecasting 2021, 3, 276–289. https://doi.org/10.3390/forecast3020018 https://www.mdpi.com/journal/forecasting