Global warming remains one of the greatest threats to the sustainability of the planet. The literature is replete with studies investigating the economic effects of global warming with very little attention being paid to the financial ramifications. To this end, the present study attempts to ascertain the predictive power of global warming for green financial assets (Clean Energy Index, Green Bond Index, World ESG Index, and Sustainability World Index) under bearish, normal and bullish market conditions. Employing a novel rolling windows wavelet quantile Granger causality testing procedure, which controls for time, frequency, and quantile asymmetries, findings reveal that the predictive power of global warming for green financial assets is more (less) stable across time at lower (higher) frequencies when markets are normal. In bearish and bullish markets, however, the predictability of global warming for green financial assets is observed to be more stable at higher frequencies and less stable at relatively lower frequencies. These results imply that global warming encourages low-frequency trading in normal markets, but induces relatively more speculative trading in bearish and bullish markets. Based on these findings, policy commendations are offered.