This study compares two causal methods, Liang–Kleeman information flow (LKIF) and Peter and Clark momentary conditional independence (PCMCI), in the context of climate analyses. The methods are applied to four artificial models of increasing complexity and one real-world case study based on climate indices in the Atlantic and Pacific regions. Both methods are shown to be superior to classical correlation analysis, especially in removing spurious links. LKIF performs better with a smaller number of variables, while PCMCI is better with a larger number of variables. Detecting causal links in the fourth model is more challenging due to its nonlinear and chaotic nature. For the real-world case study, LKIF identifies the Arctic Oscillation (AO) as the largest driver, while PCMCI identifies the El Niño–Southern Oscillation (ENSO) as the main influencing variable. The study highlights the strengths and weaknesses of both methods and suggests that further research is needed to confirm these links, particularly using nonlinear causal methods. The results demonstrate that causal methods can accurately identify true causal links and distinguish them from spurious relationships, which is crucial for understanding climate dynamics.This study compares two causal methods, Liang–Kleeman information flow (LKIF) and Peter and Clark momentary conditional independence (PCMCI), in the context of climate analyses. The methods are applied to four artificial models of increasing complexity and one real-world case study based on climate indices in the Atlantic and Pacific regions. Both methods are shown to be superior to classical correlation analysis, especially in removing spurious links. LKIF performs better with a smaller number of variables, while PCMCI is better with a larger number of variables. Detecting causal links in the fourth model is more challenging due to its nonlinear and chaotic nature. For the real-world case study, LKIF identifies the Arctic Oscillation (AO) as the largest driver, while PCMCI identifies the El Niño–Southern Oscillation (ENSO) as the main influencing variable. The study highlights the strengths and weaknesses of both methods and suggests that further research is needed to confirm these links, particularly using nonlinear causal methods. The results demonstrate that causal methods can accurately identify true causal links and distinguish them from spurious relationships, which is crucial for understanding climate dynamics.