Sensitivity and uncertainty of the carbon balance of a Pacific Northwest Douglas-fir forest during an El Nino La Nina cycle

The annual net ecosystem productivity (F-NEP) Of a second-growth Douglas-fir stand established in 1949 on the Canadian West Coast varied considerably over the 4-year period between 1998 and 2001. This period included the El Nino/La Nina cycle during the northern hemispheric winters of 1997/1998 and 1998/1999, offering a unique opportunity to study how a typical forest ecosystem in the Pacific Northwest reacts to interannual climate variability. This was possible even though annual FNEP values calculated from eddy covariance (EC) measurements of CO2 fluxes were subject to biases. These were largely due to the failure of the EC method to accurately measure losses of CO2 under low turbulence conditions at night, which caused FNEP overestimates of as much as 90 g C m(-2) per year. As these biases were largely unaffected by interannual climate variability, it was possible to reliably quantify interannual differences in F-NEP estimates if they were larger than random variability, which was estimated to be +/-30 g C m(-2) per year at most. Interannual differences were mainly due to differences in ecosystem respiration (R) between the 4 years. In the year following the 1997/1998 El Nino, high air temperatures led to the highest annual R of the 4 years, while annual gross ecosystem photosynthesis (P) was only slightly higher than normal. This resulted in 1998 having the lowest F-NEP (270 g C m(-2) per year) of the 4 years. For 1999, a cool and cloudy La Nina year, F-NEP was 360 g C m(-2) per year, much higher than 1998, but somewhat lower than the last 2 years, for which F-NEP values were 390 and 420 g C m(-2) per year, respectively. (C) 2004 Elsevier B.V. All rights reserved.