Wood is a remarkable material with unique properties, and accounts for c.60% of all living biomass. It is responsible for sequestering c.18% of fossil fuel carbon emissions annually, reducing the growth rate of atmospheric CO2 and hence mitigating climate change. While there have been many investigations into the factors controlling wood formation, which have quantified patterns and relationships, our knowledge of the underlying physiological processes is very poor. This greatly limits our understanding of the global carbon cycle, climatic and atmospheric CO2 controls on wood anatomy, and the interpretation of past environments using tree rings.

Wood formation occurs through the production of new cells just under the bark, which then undergo enlargement and wall thickening, before dying and becoming functioning xylem, the main tissue responsible for water transport and structural support. This project aims to uncover the controls, currently largely unknown, on wood formation, and hence final wood anatomy and carbon content. We will do this by measuring the effects of experimental manipulations on a wide range of anatomical, biochemical, and genomic characteristics in a clonal tree model, hybrid poplar, and use this knowledge to produce a major advance in our understanding of terrestrial carbon cycling through the development of mechanistic models of wood formation and whole-tree growth.

Dynamic Global Vegetation Models (DGVMs) are our main tools for studying global terrestrial carbon dynamics, but do not explicitly consider growth. DGVMs are instead constructed around the paradigm that plant growth is equivalent to the balance of photosynthesis and respiration, whereas there is considerable evidence that growth is controlled independently of the overall carbon balance. If true, this would largely invalidate the basic premise of these models. The limitations of the current approach are evident in the finding that while DGVMs, on average, reproduce the magnitude of the historical terrestrial carbon sink, they have very different underlying climate and CO2 sensitivities and therefore cannot be considered to provide a mechanistic explanation of the processes behind this imbalance. A growth-led DGVM has the potential to resolve this problem and thereby greatly improve our ability to predict the behaviour of the future global carbon cycle. This has so far not been possible due to the lack of fundamental physiological knowledge. Our experimental manipulations on hybrid poplar will produce the knowledge necessary for the incorporation of explicit growth into DGVMs, overturning their most fundamental paradigm, and so enable a major breakthrough in understanding and predictability.

Our highly innovative collaboration brings together a research group working at the cutting edge of the molecular physiology of wood formation, with one highly experienced in the development of plant growth models and DGVMs. This unique collaboration has the potential to produce a major advance in our understanding of wood formation directly targeted at the needs of terrestrial carbon cycling models.

The abiotic factors to be considered are temperature, atmospheric CO2, daylength, and soil moisture. We will investigate how carbohydrate supply interacts with temperature in controlling cell numbers, sizes, and wall thicknesses. In another series of experiments, we will produce growth rings under an accelerated annual growth cycle using changing daylength and temperature, and analyse how the influences of these factors on ring anatomy are achieved through changes in the rates and durations of differentiation phases. Soil moisture manipulations will be used to quantify the influence of drought on wood anatomy through effects on cell enlargement, proliferation, and carbohydrate supply, and hybrid poplar will be grown under field conditions for the entire length of the project to provide a strong test of our new understanding of tree growth.