Interannual variation of soil respiration in a beech forest ecosystem over a six-year study

Soil respiration was measured for six years from June 1996 to December 2001 in order to investigate both seasonal and interannual variations in a young beech forest in North Eastern France (Hesse forest). Soil respiration exhibited pronounced seasonal variations that clearly followed the seasonal changes in soil temperature (T) and soil water content (W). An exponential function (y = AWe(BT)) fitted the data well, and including a linear effect of soil water content on soil respiration strongly improved the predictive capacity of the model. However, the increase in residuals when plotted against the date of measurements clearly evidenced that changes in soil temperature or soil water content failed to predict the increase in soil respiration with time, highlighting that the interannual variation in soil respiration was not solely due to direct climate effects. When fitted against single year data, the temperature sensitivity coefficient ( B) was very close for both years while the pre-exponential factor ( A) for 1997 was half of those for 2001. The model was further run over the entire data set, allowing A to vary from one year to another. There was a close agreement between predicted and observed soil respiration and A exhibited a linear trend with time with a high value for 1999 after thinning.