COS 32-4
Phenological responses of a temperate forest understory to experimental warming
Increasing temperature has the potential to affect vegetative and reproductive phenology of temperate forest ecosystems in the future. The impact of warming on temperate forest communities is highly uncertain in part because warming experiments have focused on open ecosystems such as tundra, alpine meadows, and grasslands. Recent findings have also indicated that many warming experiments underpredict plant responses to climate change. Here we took advantage of an ongoing warming experiment to understand the influence of temperature on phenology of a diverse suite of 11 temperate forest species known to differ in their temporal phenological niche and xylem anatomy. Our primary goal was to quantify how forest species alter budburst, flowering time, and autumn leaf coloring in response to warming of 1.5–5.5 °C. Leaf and flower phenology was measured inside actively heated open-top chambers in the understory of an oak-hickory forest in Duke Forest, North Carolina. To determine if this warming experiment underpredicted plant responses to climate change, we compared the effect of experimental warming on budburst and flowering date to interannual differences in phenology observations between 2011 and 2013, two years when mean spring temperature differed by about 3 °C.
Results/Conclusions
Warming advanced budburst of six species by 0.8–3.5 days per °C and delayed leaf coloring in autumn by 3.1–5.8 days per °C, resulting in an extension of the 2011 growing season by 3.7–6.3 days per °C. We found nonlinear responses of budburst phenology to strong warming, where shifts in budburst (per °C) were larger for warming of 1–2 °C than for 3–5 °C. Advances in budburst were larger in diffuse-porous than ring-porous species, possibly due to more conservative safety mechanisms in ring-porous species for prevention of damage by late-spring frosts. Warming advanced flowering of two species by 3.7 and 10.5 days per °C, delayed flowering of one species by 6.1 days per °C, and had no effect on four species. Experimental warming in the spring appeared to increase developmental rates of plant species, whereas warming in mid-summer exceeded optimal temperatures for reproduction. Observations of interannual temperature variability overpredicted advancement in budburst by 3-fold, relative to our warming experiment, and failed to consistently predict the direction of shifts in flowering time. These results suggest that high rates of phenological change observed in the past are unsustainable and will decrease with warming throughout the coming century.