To establish relationships between ecosystem photosynthesis (PG) and evapotranspiration, PG was calculated as the sum of net ecosystem productivity (FNEP) and ecosystem respiration (RE), i.e. PG = FNEP + RE. Daytime RE was calculated from the relationship of soil temperature and nighttime FNEP = - RE, which was allowed to vary seasonally. This estimate usually agrees to within 10% with RE estimated as the intercept of the light response of FNEP. Nighttime measurements were only used for the analysis when friction velocity was above a threshold of 0.3 m/s. The pattern of interannual variability did not vary with either this threshold or the use of an annual instead of a seasonal soil temperature relationship.
Analysis of monthly light response of PG showed that maximum assimilation rates at high light levels varied from 10 μmol m-2 s-1 in December and January to just under 30 μmol m-2 s-1 in July and August. Low soil moisture in late summer led to the reduction of maximum assimilation. However, monthly totals of PG were related linearly to downwelling photosynthetically active radiation and there was little interannual variation in the light use efficiency, which had an overall value of 0.25 g C / mol photons. Similarly, stomatal conductance tended to decrease with decreasing soil moisture, yet this was only important during late summer. Water use efficiency was also roughly constant seasonally as well as for different years. Using monthly values, the stand sequestered 6 g C / kg H2O with monthly PG as high as 380 g C m-2 in summer. The 1997/1998 El Niño event led to an unusually warm spring in 1998, leading to an increase in respiration and consequently low annual FNEP of 301 g C m-2. The 2002/2003 El Niño caused an unusually mild early winter in 2002 which again caused an increase in RE and an annual FNEP of 206 g C m-2.