4. Discussion
Here we have assessed the two most critical transport systems in trees, xylem in sapwood and phloem in bark, as one tightly interconnected system. This approach is supported by previous experimental work that highlighted the role of wood rays as a radial link between the two transport systems (Pfautsch et al. 2015b; Mencuccini et al. 2017). By refining our Stem Hydraulic Model (Mencuccini et al. 2017) through incorporation of terms for viscoelastic changes driven by pressure potential of the xylem and osmotic potential exerted by live bark, we were able to accurately isolate diel patterns of irreversible radial growth (D G+) from asynchronous dimensional changes in bark and sapwood. The resulting patterns of D G+ showed high rates of wood growth during the day, which stand in contrast to the paradigm that radial growth of stems mostly occurs during the night (Steppe et al. 2015). Moreover, by girdling trees, we were able to untangle the effects of hydraulic and osmotic pressure gradients on radial movement of bark and sapwood. Our observed and modelled dimensional changes of tree stems progress recent theoretical work on source-sink relationships (Sellier & Mammeri 2019). Specifically, we were able to quantify dimensional radial expansion and contraction of tissues driven by diurnal oscillations of osmotic pressure potential in live bark and xylem. Further, we identified that D G+ close to the source of loading (treetop) followed a pronounced diel pattern of high growth during the day and low growth during the night, compared to continuous increase of D G+ further away from the source of loading (base). These growth patterns have now been shown to co-occur simultaneously in eucalypt trees, providing the first empirical evidence for the suggested pathway effects of phloem loading with a stabilising function of NSC supply to distant sinks (Sellier & Mammeri 2019).