Respiration characteristics in temperate rainforest tree species differ along a long-term soil-development chronosequence

We measured the response of dark respiration (R d) to temperature and foliage characteristics in the upper canopies of tree species in temperate rainforest communites in New Zealand along a soil chronosequence (six sites from 6 years to 120,000 years). The chronosequence provided a vegetation gradient characterised by significant changes in soil nutrition. This enabled us to examine the extent to which changes in dark respiration can be applied across forest biomes and the utility of scaling rules in whole-canopy carbon modelling. The response of respiration to temperature in the dominant tree species differed significantly between sites along the sequence. This involved changes in both R d at a reference temperature (R 10) and the extent to which R d increased with temperature (described by E o, a parameter related to the energy of activation, or the change in R d over a 10°C range, Q 10). Site averaged E o ranged from 44.4 kJ mol -1 K -1 at the 60-year-old site to 26.0 kJ mol -1 K -1 at the oldest, most nutrient poor, site. Relationships between respiratory and foliage characteristics indicated that both the temperature response of respiration (E o or Q 10) and the instantaneous rate of respiration increased with both foliar nitrogen and phosphorus content. The ratio of photosynthetic capacity (Whitehead et al. in Oecologia 2005) to respiration (A max/R d) attained values in excess of 15 for species in the 6- to 120-year-old sites, but thereafter decreased significantly to around five at the 120,000-year-old site. This indicates that shoot carbon acquisition is regulated by nutrient limitations in the retrogressing ecosystems on the oldest sites. Our findings indicate that respiration and its temperature response will vary according to soil age and, therefore, to soil nutrient availability and the stage of forest development. Thus, variability in respiratory characteristics for canopies should be considered when using models to integrate respiration at large spatial scales.