Productivity and water use of Australian tree species under climate change

Climate change is likely to impact heavily on Australian forests. Over the next century Australian forests will likely experience an increase in the frequency and severity of drought conditions, and an increase in temperature. Such changing climate conditions may severely disrupt the role of Australian forests in regional and global carbon and water cycles. However, there are considerable gaps in our understanding of the impacts of these changes in climate on forests. In this thesis I examine two key questions for Australian forest responses to climate change. Firstly, it is commonly hypothesized that elevated CO2 will ameliorate the impact of drought on forest growth and productivity, but there is little evidence to date to support this hypothesis. To address this gap, I examine the impact of elevated CO2 and variable drought conditions on two Eucalyptus species of contrasting drought tolerance. Secondly, we have little understanding of the effect of increasing temperature on leaf level physiology in species from warm climates. To address this gap, I examine the impact of temperature on leaf level physiology of two Australian tropical rainforest species. It is often hypothesized that elevated CO2 will impart the greatest relative benefit to forest ecosystems under water limitation, and therefore that elevated CO2 will reduce the impact of drought. There are two main mechanisms underlying this hypothesis. The first is that lower intercellular CO2 (Ci) occurring under drought conditions causes a larger enhancement of photosynthesis (A) relative to atmospheric CO2 concentration (Ca) due to the non-linear response of A to Ci. Further, the higher slope of the A-Ci curve at lower Ca leads to a greater reduction of A, due to drought, under ambient than elevated Ca. Secondly, stomatal conductance is reduced under elevated CO2, lowering transpiration rate and conserving soil moisture, thus enabling trees under elevated CO2 to continue to transpire longer into a drought episode. In this thesis, I explore the two mechanisms for CO2 x drought interactions separately, using a large experiment with two Eucalyptus species of contrasting drought tolerance in elevated (700 ppm) and ambient (380 ppm) CO2 glasshouses. To explore the first mechanism - lower intercellular CO2 under drought conditions - seedlings of mesic Eucalyptus pilularis and xeric Eucalyptus populnea were grown at soil moisture content of either 50% or 100% of field capacity (FC) for 9-11 months. We hypothesized that water-use efficiency (ratio of carbon gain to water loss) would be proportional to growth CO2 (i.e. in this experiment, elevated CO2 would cause an increase in WUE of 84%, the ratio of elevated (700 ppm) to ambient CO2 (380 ppm)). We hypothesised that this increase would be the same for both drought treatments and species, but that lower Ci in the droughted plants and xeric species would lead to relatively larger CO2 effects on photosynthesis and biomass growth, and smaller CO2 effects on transpiration, than in the well-watered plants and the mesic species. These hypotheses were rejected. At the leaf level, instantaneous transpiration efficiency (ratio of photosynthesis to transpiration) responded more than proportionally to growth CO2 (i.e. greater than 84%). Leaf gas exchange was not affected by growth under long-term moderate drought and did not differ between species, thus leading to a rejection of the low Ci mechanism. At whole plant scale, the CO2 effect on whole-plant water use efficiency (WUE) was considerably less than the increase in CO2. For both species, transpiration rate was similar for plants grown under elevated or ambient CO2, reflecting an increase in leaf area to compensate for the CO2-induced reduction in gs. These results suggest that under elevated CO2 and long-term moderate drought both mesic E. pilularis and xeric E. populnea exhibit a capacity to adjust growth processes to match water availability in order to avoid moderate drought stress. A test of the second proposed mechanism (soil water savings under elevated CO2), was carried out at the end of the large experiment. This test involved bringing all pots of E. pilularis and E. populnea back to full field capacity and allowing plants to dry down to predetermined physiological stress levels. There were clear differences among species and antecedent watering treatments in the effect of CO2 on water stress (identified as a change from the maximum in photosynthesis (A) and stomatal conductance (gs) observed under well-watered conditions, when pre-dawn leaf water potential (Ψpd) was close to zero). A delay in water stress in this context represents a longer period of time taken to reach a minimum in A and gs from well-watered conditions. During the dry down experiment, elevated CO2 strongly delayed water stress in well-watered E. populnea but had no effect on progress of water stress in well-watered E. pilularis. Plants of both species grown under low water availability showed some reduction in water stress with elevated CO2. These responses can be understood from the perspective of individual species ecological strategies: under well-watered conditions E. pilularis grows rapidly in response to elevated CO2, making it vulnerable to future drought, whereas E. populnea responds conservatively to elevated CO2, allowing soil moisture savings when drought occurs. Lastly, a better understanding of the effect of warming on leaf level physiology for tropical rainforests is needed to assist with better parameterization of global scale models of forest response to climate change. In this experiment I measured the biochemical and stomatal limitations to leaf level photosynthesis in response to temperature on two canopy species at the Australian Canopy Crane Research Station (ACCRS) in Cape Tribulation, Queensland, Australia. Data were used to perform a sensitivity analysis of a coupled photosynthesis-stomatal model, comparing rainforest parameter values with two parameter sets commonly used for modelling from Leuning (2002) and Kattge and Knorr (2007). The analysis showed that general parameters for C3 species used in global scale models of forest responses to climate change under predict the optimum temperature of photosynthesis for tropical forest species, even when temperature acclimation is taken into account. The parameter values obtained in this study will prove useful for improving global vegetation models.