Projected timing of perceivable changes in climate extremes for terrestrial and marine ecosystems

Human and natural systems have adapted to and evolved within historical climatic conditions. Anthropogenic climate change has the potential to alter these conditions such that onset of unprecedented climatic extremes will outpace evolutionary and adaptive capabilities. To assess whether and when future climate extremes exceed their historical windows of variability within impact‐relevant socioeconomic, geopolitical, and ecological domains, we investigate the timing of perceivable changes (time of emergence; TOE) for 18 magnitude‐, frequency‐, and severity‐based extreme temperature (10) and precipitation (8) indices using both multimodel and single‐model multirealization ensembles. Under a high‐emission scenario, we find that the signal of frequency‐ and severity‐based temperature extremes is projected to rise above historical noise earliest in midlatitudes, whereas magnitude‐based temperature extremes emerge first in low and high latitudes. Precipitation extremes demonstrate different emergence patterns, with severity‐based indices first emerging over midlatitudes, and magnitude‐ and frequency‐based indices emerging earliest in low and high latitudes. Applied to impact‐relevant domains, simulated TOE patterns suggest (a) unprecedented consecutive dry day occurrence in >50% of 14 terrestrial biomes and 12 marine realms prior to 2100, (b) earlier perceivable changes in climate extremes in countries with lower per capita GDP, and (c) emergence of severe and frequent heat extremes well‐before 2030 for the 590 most populous urban centers. Elucidating extreme‐metric and domain‐type TOE heterogeneities highlights the challenges adaptation planners face in confronting the consequences of elevated twenty‐first century radiative forcing.