An experimental test of the causes of forest growth decline with stand age

The decline in aboveground wood production after canopy closure in even-aged forest stands is a common pattern in forests, but clear evidence for the mechanism causing the decline is lacking. The problem is fundamental to forest biology, commercial forestry (the decline sets the rotation age), and to carbon storage in forests. We tested three hypotheses about mechanisms causing the decline in wood growth by quantifying the complete carbon budget of developing stands for over six years (a full rotation) in replicated plantations of Eucalyptus saligna near Pepeekeo, Hawaii. Our first hypothesis was that gross primary production (GPP) does not decline with stand age, and that the decline in wood growth results from a shift in partitioning from wood production to respiration (as tree biomass accumulates), total belowground carbon allocation (as a result of declining soil nutrient supply), or some combination of these or other sinks. An alternative hypothesis was that GPP declines with stand age and that the decline in aboveground wood production is proportional to the decline in GPP. A decline in GPP could be driven by reduced canopy leaf area and photosynthetic capacity resulting from increasing nutrient limitation, increased abrasion between tree canopies. lower turgor pressure to drive foliar expansion, or hydraulic limitation of water flux as tree height increases. A final hypothesis was a combination of the first two: GPP declinus, but the decline in wood production is disproportionately larger because partitioning shifts as well. We measured the entire annual carbon budget (aboveground production and respiration, total belowground carbon allocation [TBCA], and GPP) from 0.5 years after seedling planting through 61/2, years (when trees were similar to25 in tall). The replicated plots included two densities of trees (1111 trees/ha and 10 000 trees/ha) to vary the ratio of canopy leaf mass to wood mass in the individual trees, and three fertilization regimes (minimal, intensive, and minimal followed by intensive after three years) to assess the role of nutrition in shaping the decline in GPP and aboveground wood production. The forest closed its canopy in 1-2 years, with peak aboveground wood production, coinciding with canopy closure, of 1.2-1.8 kg C(.)m(-2.)yr(-1). Aboveground wood production declined from 1.4 yr(-1.) at age 2 to 0.60 kg C-m(-2). kg, C(.)m(-2.)yr(-1) at a yr(-1) at age 6. Hypothesis I failed: GPP declined from 5.0 kg C(.)m(-2.)yr(-1) at age 2 to 3.2 kg C(.)m(-2 .)yr(-1) at age 6. Aboveground woody respiration declined from 0.66 kg C(.)m(-2.)yr(-1) at age 2 to 0.22 kg C.m(-2.) yr(-1) at age 6 and TBCA declined from 1.9 kg C(.)m(-2.)yr\(-1) C, C, at age 2 to 1.4 kg C(.)m(-2.)yr(-1) at age 6. Our data supported hypothesis 3: the decline in aboveground wood production (42% of peak) was proportionally greater than the decline in canopy photosynthesis (64% of peak). The fraction of GPP partitioned to belowground allocation and foliar respiration increased with stand age and contributed to the decline in aboveground wood production. The decline in GPP was not caused by nutrient limitation, a decline in leaf area or in photosynthetic capacity, or (from a related study on the same site) by hydraulic limitation. Nutrition did interact with the decline in GPP and aboveground wood production, because treatments with high nutrient availability declined more slowly than did our control treatment, which was fertilized only during stand establishment.