Evaluation of Data-Driven Causality Discovery Approaches among Dominant Climate Modes

• Main Authors: Hussung, Steve, Mahmud, Suhail, Sampath, Akila, Wu, Mengxi, Guo, Pei
• Format: Article in Journal/Newspaper
• Language: unknown
• Published: HPCF UMBC 2019
• Subjects: causal networks; atmospheric teleconnection patterns; data-driven causal discovery methods; UMBC High Performance Computing Facility HPCF; Pacific; North Atlantic; North Atlantic oscillation
• Online Access: https://dx.doi.org/10.13016/m2arfa-fzkx; http://mdsoar.org/handle/11603/16929

Research Assistant: Pei Guo Faculty Mentor: Jianwu Wang : Identification of causal networks in atmospheric teleconnection patterns has applications in many climate studies. We evaluate and compare three data-driven causal discovery methods in locating and linking causation of well-known climatic oscillations. Four climate variables in the ERA-Interim reanalysis data (1979-2018) were examined in the study. We first employ dimension reduction to derive the the time-series for selected climate variables. Then timeseries of dominant modes were processed using three different causal discovery methods: Granger causality discovery, Convergent cross-mapping (CCM), and PCMCI. Discovered causal links were different for different methods as well as for different variables. However, slightly similar causal links were observed between the Granger causality and CCM methods. Comparison of these three methods is discussed based on the El Ni˜no-Southern Oscillation (ENSO) and its connection with other oscillations. Causal discovery methods were able to capture the linkage between the ENSO, North Atlantic Oscillation (NAO), and Pacific Decadal Oscillation (PDO), for some of the variables. Overall, this study identifies the usage of these statistical models in locating the direct and indirect causal links among the oscillations. Application of these data-driven causal discovery methods, both in terms of mediation and direct relationships between the observed teleconnection patterns, suggests that the data-driven statistical methods are efficient in locating the regimes of climate patterns and their 12 observed real connections to some extent. We present and provide our explanation of the evaluation results for each of the three causal discovery methods.