95th ESA Annual Meeting (August 1 -- 6, 2010)

PS 48-177 - CANCELLED - Regime shift in gross ecosystem exchange at Harvard Forest

Tuesday, August 3, 2010
Exhibit Hall A, David L Lawrence Convention Center
Lingyun Wu, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing, China and Jingyong Zhang, School of Forestry and Wildlife Sciences, Auburn University
Background/Question/Methods

Ecosystems usually respond to gradual climate changes in a smooth way. However, the smooth way change can be interrupted by sudden drastic switches to a contrasting alternative stable state. Climate regime shift described using Pacific Decadal Oscillation (PDO) can lead to abrupt changes in temperature, precipitation, and solar radiation, which dominate the interannual fluctuation in terrestrial ecosystem CO2 exchange. Long-lived and relatively immobile unmanaged ecosystems such as mature forest are likely to be especially sensitive to abrupt climate change. The most recent climate regime shift of 1998-1999 has been documented on the basis of physical and biological information. The new climate shift is associated with a transition from El Niño (1998) to La Niña (1999) that coincides with a PDO reversal from a warm to a cool phase. Although catastrophic climate change connected with the shift has been investigated, the response of terrestrial ecosystem CO2 exchange has not yet been identified. In this study, we use 1992-2007 records of CO2 flux, measured in an 80-115-year-old deciduous forest in Central Massachusetts, to examine change in carbon exchange associated with the climate shift. An automatic sequential algorithm is applied to detect regime shifts in carbon exchange in the forest.

Results/Conclusions

Our results show that gross ecosystem exchange (GEE) at Harvard Forest appears to have changed substantially during 1992 to 2007 in coincidence with the climate regime shift that occurred around 1998-1999. The automatic sequential algorithm detects that this event can be termed “regime shift”. The GEE conspicuously jumped from one rather stable condition (1992-1998) to another (2001-2007), with 1999 and 2000 appearing as “transition” years. The GEE regime shift correlates well with 1998-1999 climate regime shift at a lag of about 2 years. The GEE shift is not explained well by land-use change and disturbance, and changes in nitrogen deposition and CO2 fertilization. The results further suggest that water and surface air temperature have little effect on the GEE regime shift. Instead, photosynthetically active radiation (PAR) significantly increases after GEE regime shift and shows a pronounced regime shift since 2001 concurrent with GEE shift. This suggests that changes in the light can reasonably be expected to be a driver of the shift.