It is the timing of climatic extremes that determines their impact on carbon cycling

Changes in climatic conditions and their effects on phenology and carbon cycling of terrestrial ecosystems are well documented at local to global scales. While there is a general consensus on the existence of such links between warming trends, changes in phenology (e.g. vegetative season length) and the effectiveness of ecosystems as carbon sinks, the impacts of short-term climatic extreme events (e.g., droughts or frosts) on carbon cycling, in combination with long-term trends, are less certain. Here, we argue that the timing of extreme events relative to phenology determines whether, and how, net ecosystem exchange of carbon (NEE) is affected. Further, we hypothesize that long-term trends in phenology and climate make forest ecosystems more susceptible to some extreme events, and less so to others.

In a recent study, Dragoni et al. (2010, Global Change Biology 17, 886-897) reported on long-term trends in carbon exchange at the Morgan Monroe State Forest in Indiana (MMSF, USA). From 1998 to 2008, an overall increase in late-summer temperatures led to an extension of the active vegetative season progressively later into the fall and was shown to drive significantly increased net ecosystem productivity (NEP) at MMSF (enhanced carbon uptake responsible for an increase of NEP by about 50 gC m-2 over a decade). However, in later years (2009 to 2012), a series of extreme short-term climate events, ranging from early to late season droughts, or colder-than-normal springs, have led to very low annual productivity (on average 25% lower than the 1998-2008 mean). Such extreme climate events impacted phenology, carbon exchange and allocation, and nutrient cycling. In part due to the fall extension of the carbon uptake season, the 2012 drought started already in the fully active uptake period and led to strong declines in NEE that canceled out the effects of an early start to the season. Late summer droughts in earlier years started close to or after the onset of senescence and had considerably smaller effects on annual NEP. Thus, depending on their timing, extreme climate events have the potential to outweigh positive trends in NEP that are induced by gradual shifts in climate and phenology. Given that extreme short-term climate events are expected to increase in intensity and frequency in a warming world, our results highlight the need to investigate the role of short-term climatic stressors in the acclimation of ecosystems to gradual shifts in climate.