Abstract
Episodically occurring internal (climatic) and external (non-climatic) disruptions of normal climate variability are known to both affect spatio-temporal patterns of global surface air temperatures (SAT) at time-scales between multiple weeks and several years. The magnitude and spatial manifestation of the corresponding effects depend strongly on the specific type of perturbation and may range from weak spatially coherent yet regionally confined trends to a global reorganization of co-variability due to the excitation or inhibition of certain large-scale teleconnectivity patterns. Here, we employ functional climate network analysis to distinguish qualitatively the global climate responses to different phases of the El Niño–Southern Oscillation (ENSO) from those to the three largest volcanic eruptions since the mid-20th century as the two most prominent types of recurrent climate disruptions. Our results confirm that strong ENSO episodes can cause a temporary breakdown of the normal hierarchical organization of the global SAT field, which is characterized by the simultaneous emergence of consistent regional temperature trends and strong teleconnections. By contrast, the most recent strong volcanic eruptions exhibited primarily regional effects rather than triggering additional long-range teleconnections that would not have been present otherwise. By relying on several complementary network characteristics, our results contribute to a better understanding of climate network properties by differentiating between climate variability reorganization mechanisms associated with internal variability versus such triggered by non-climatic abrupt and localized perturbations.
Highlights
While the analysis of such big climate data sets has been traditionally attempted by means of classical statistical approaches like empirical orthogonal function (EOF) or maximum covariance analysis [4], it has recently been realized that these methods exhibit fundamental intrinsic limitations, including their linearity and associated condition of pairwise orthogonal patterns (e.g. [5])
We present the results of our functional network analysis of global surface air temperature (SAT) patterns with a focus on the associated imprints of El Nino–Southern Oscillation (ENSO)
Following upon the latter result, we expect that the time dependence of the correlation threshold does not add much complementary information that is not provided by the corresponding behavior of the network transitivity
Summary
While the analysis of such big climate data sets has been traditionally attempted by means of classical statistical approaches like empirical orthogonal function (EOF) or maximum covariance analysis [4], it has recently been realized that these methods exhibit fundamental intrinsic limitations, including their linearity and associated condition of pairwise orthogonal patterns (e.g. [5]). One specific example of such methodological developments are complex network representations of climate variability [11,12,13,14,15,16,17,18,19]. They share some common roots with classical techniques like EOF analysis, they generalize the corresponding scope and can potentially relieve some of the aforementioned concerns [20]. We focus on the so-called functional climate network analysis, in which the individual grid points or cells are considered as nodes of a spatially embedded graph. Related to the aforementioned line of research are approaches attempting to define climate areas, i.e., regions with coherent climate variability, based on spatial clustering of grid cells in a fully connected weighted climate network [38] or other multivariate analysis techniques [10,39]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
