Representing Nutrition of Pinus Radiata in Physiological Hybrid Productivity Models
Degree GrantorUniversity of Canterbury
Degree NameDoctor of Philosophy
Hybrid physiological models are being increasingly used to assess productivity, carbon sequestration, water and nutrient use and environmental impacts of management decisions. Users include forest managers, politicians, environmental agencies and scientists. However a wider use of these models has been prevented as a result of an incomplete understanding of the mechanisms regulating carbon allocation, nutrient availability in soils and nutrient uptake by trees. On-going innovation in clonal forestry, genetic improvement and vegetation management techniques is also poorly represented in hybrid models. This thesis examines means to represent nutrition and genotype-nutrition interactions in productivity physiological hybrid models. Nutrient limitations and growth differences between genotypes were hypothesized to operate through key physiological processes: photosynthesis, carbon allocation and nutrient internal cycling. In order to accomplish the aims of the study both greenhouse and field experimentation were carried out. In a first experiment, responses of photosynthesis (A) to intercellular CO₂ concentration (Ci) were measured in a fast- and a slow-growing clone of Pinus radiata D. Don cultivated in a greenhouse in a factorial combination of nitrogen and phosphorus supply, and analyzed using the biochemical model of leaf photosynthesis described by Farquhar et al. (1980). There were significant positive linear relationships between the parameters, Vcmax, Jmax, Tp and both foliar nitrogen (Na) and phosphorus (Pa) concentration on an area basis. The study showed that the effects of nitrogen and phosphorus supply on photosynthesis were statistically independent and that the photosynthetic behaviour of the two clones was equivalent. In a similar study, gas exchange and chlorophyll fluorescence were simultaneously measured to determine internal transfer conductance (gm) based on the "constant J method". Transfer conductance may pose significant limitations to photosynthesis which may be differentially affected by nutrition and genotype in Pinus radiata. Values of gm were similar to those of stomatal conductance (gs) and their ratio (gm / gs) was not influenced by nutrient supply or clone being on average (±1 SE) 1.22 ±0.04. Relative mesophyll limitations (LM, 16%) to photosynthesis were marginally greater than those imposed by stomata (LS, 13%), and together smaller than the relative limitations posed to photosynthesis by biochemical processes (LB, 71%). The CO₂ concentration in the intercellular air spaces (Ci) was (±1 SE) 53 ±3 µmol mol-1 lower than in the atmosphere (Ca) while CO₂ concentration in the chloroplasts (Cc) was (±1 SE) 48 ±2 µmol mol-1 less than Ci. Values of LS, LM and LB and CO₂ diffusion gradients posed by gs (Ca-Ci) and gm (Ci-Cc) did not change with nutrient supply or clone. In a third experiment, one-year old Pinus radiata cuttings from four genotypes were cultivated in silica sand with a factorial combination of nitrogen (N0=1.43 and N1=7.14 mM) and phosphorus (P0=0.084 and P1=0.420 mM) supply for 24 months. N supply was enriched with ¹⁵N to 2.5⁰/₀₀ (labelled N) during the first year, then plants transferred to clean sand and cultivated for another year with ¹⁵N at levels close to natural abundance (0.3664899 atom percent ¹⁵N, δ¹⁵N 0.5115 ⁰/₀₀) provided by the source of N in nutrient solution applied during the second year. Recovery of labelled and unlabelled N was used to estimate N remobilization. N remobilization scaled with plant growth, N content and N and P supply. In relative terms, 65% of all stored N was remobilized in the high-nutrient supply regime compared to 42-48% at lower N and P addition rates. Most N remobilization occurred during spring-summer (77%), coincidently with the largest proportion of needle development (80%), indicating that N remobilization was driven by sink-strength. Foliage was by far the main source for internal cycling while roots were the main sink (40%). Clones exhibited differences in N remobilization capacity, but these differences were completely explained by the size of the N pool before remobilization took place, indicating that N remobilization performance was similar among clones. In a fourth study, four clones were cultivated in silica sand with a factorial combination of nitrogen and phosphorus supply for ten months, and patterns of carbon allocation examined using a carbon balance approach. Gross-primary productivity (GPP) scaled mainly with nitrogen but also with phosphorus supply. The fraction of GPP (GPP = ANPP + APR + TBCA) allocated to above-ground components (ANPP) increased with N and P supply at the expense of total-below ground C allocation (TBCA) with no apparent effect on the fraction of GPP partitioned to above-ground plant respiration (APR). Carbon use efficiency (NPP:GPP) scaled with nutrient supply, being 0.42 in the low-nutrient supply regime compared to 0.51 in the high-nutrient supply regime, suggesting that in poor fertility environments a larger proportion of the C budget is respired compared to the net productivity. Fast-growing clones allocated about 2-4% more carbon to above-ground components (ANPP) at the expense of carbon allocated below-ground (TBCA) with no effect on carbon respired above-ground (APR), indicating that faster-growing genotypes allocate more carbon to leaf area which may compound and increase overall GPP over time. The field component of this thesis was conducted in a subset of locations where ENSIS (formerly New Zealand Forest Research Institute) had established trials to test the influence of species, soil disturbance and plant nutrition on sustainability indicators. Plots were small in size (3 m × 3 m) with trees spaced at 0.5 m × 0.5 m (40 000 trees ha-1) with nine measurement trees surrounded by a two-row buffer. All sites were planted in winter 2001 and harvested in spring 2005. The aim of this pilot study was to examine patterns of carbon allocation during the fourth year after planting in control and fertilized mini-plots of Pinus radiata in five sites with contrasting climate and soil conditions in the South Island of New Zealand. The study showed that the fraction of gross-primary productivity allocated belowground increased as the soil C:N ratio increased. However, these results should be interpreted with caution due to the unusual nature of the trial and the reduced number of sites studied. Two existing physiological models were selected for the discussion in this thesis (3-PG, Landsberg and Waring 1997; canopy net carbon exchange model, Whitehead et al. 2002). Potential improvements for the nutritional component of 3-PG comprise: accounting for reductions in carbon use efficiency (NPP:GPP) in poor-fertility environments, adding a preliminary fertility modifier (FN, 0-1) driven by soil C : N ratio and soil N, adding a preliminary relationship between carbon allocation to roots and the soil C : N ratio and representing faster-growing genotypes by increasing their leaf area but not their photosynthetic performance. The canopy net carbon exchange model (NCE) combines the coupled model of leaf photosynthesis - stomatal conductance described by Leuning (1995) with canopy structure and a water balance model to scale carbon assimilation from leaves to canopies. Potential improvements to account for nutrient deficiencies in the leaf model by Leuning (1995), comprise using nutrient ratios to discriminate nitrogen (Na/Pa < 23 mol mol-1) from phosphorus deficiencies (Na/Pa > 23 mol mol-1), adding relationships between photosynthetic model parameters Vcmax and Jmax to Pa, and correcting the estimation of photosynthetic parameters Vcmax and Jmax by accounting for transfer conductance (gm). The canopy net carbon exchange model may be also modified to account for carbon-use efficiency, carbon allocation to roots and genotype in a similar form to that proposed for 3-PG. The results previously outlined provide a preliminary framework to represent tree and soil nutrition in physiological hybrid productivity models.