Skip to main content

Early Warning Signals & Declining Resilience of a Northern Hardwood Forest


Full title: Using Early Warning Signals of Change to Illustrate Declining Resilience of the Biology and Biogeochemistry of a Northern Hardwood Forest

Speaker: Dr. Alexandra Contosta, Research Assistant Professor, Earth Systems Research Center, University of New Hampshire

Resilience is an emergent property of an ecosystem that reflects the amount of disturbance the system can withstand before shifting into an alternative stable state. Often resilience is assessed with “lagging indicators” such as hysteresis that monitor changes in ecosystem variables in response to previous or ongoing disturbances. By contrast, “leading indicators” or “early warning signals” focus on detecting whether and how ecosystems might respond to future shocks and feature the temporal behavior of simple statistics of ecosystem variables, including coefficient of variation, autocorrelation, skewness, and kurtosis.

Prior studies that have developed and tested early warning signals of ecological resilience have occurred in highly controlled, experimental conditions, have been the product of computer simulations, or have focused on aquatic environments. Here I take advantage of the long-term (< 50 year) record of the biology and biogeochemistry at the Hubbard Brook Experimental Forest in New Hampshire, USA to explore how early warning signals of ecological resilience might illuminate a forest ecosystem’s responses to disturbance.

While previous research has shown these forests recover from a variety of disturbances, early warning signals suggest declining resilience in watershed processing of nitrogen and calcium and the viability of key tree, bird, and insect populations. Changes in early warning signals were also highly correlated with changes in precipitation chemistry and temperature. This analysis demonstrates the potential to apply a single analytical framework across a wide variety of ecosystem variables collected over decadal timescales to detect phenomena such as critical slowing down and flickering, despite natural variation and exogenous noise. It also suggests that long term shifts in climate and atmospheric deposition have reduced northern hardwood forest resilience and consequently its ability to provide important ecosystem services.