Water relations in tree physiology: where to from here?

We look back over 50 years of research into the water relations of trees, with the objective of assessing the maturity of the topic in terms of the idea of a paradigm, put forward by Kuhn in 1962. Our brief review indicates that the physical processes underlying the calculation of transpiration are well understood and accepted, and knowledge of those processes can be applied if information about the leaf area of trees, and stomatal conductance, is available. Considerable progress has been made in understanding the factors governing stomatal responses to environment, with insights into how the hydraulic conducting system of trees determines the maximum aperture of stomata. Knowledge about the maximum stomatal conductance values likely to be reached by different species, and recognition that stomatal responses to increasing atmospheric vapor pressure deficits are in fact responses to water loss from leaves, provides the basis for linking these responses to information about hydraulic conductance through soil–root–stem–branch systems. Improved understanding in these areas is being incorporated into modern models of stomatal conductance and responses to environmental conditions. There have been significant advances in understanding hydraulic pathways, including cavitation and its implications. A few studies suggest that the major resistances to water flux within trees are not in the stem but in the branches. This insight may have implications for productivity: it may be advantageous to select trees with the genetic propensity to produce short branches in stands with open canopies. Studies on the storage of water in stems have provided improved understanding of fluxes from sapwood at different levels. Water stored in the stems of large trees may provide up to 20–30% daily sap flow, but this water is likely to be replaced by inflows at night. In dry conditions transpiration by large trees may be maintained from stored water for up to a week, but flows from storage may be more important in refilling cavitated xylem elements and hence ensuring that the overall hydraulic conductivity of stems is not reduced. Hydraulic redistribution of water in the soil may make a contribution to facilitating root growth in dry soils and modifying resource availability. We conclude that the field of tree water relations is mature, in the sense that the concepts underlying models describing processes and system responses to change are well-tested and accepted and there are few, if any, serious anomalies emerging. Models are essentially formal statements about the way we think systems work. They are always subject to further testing, refinement and improvements. Gaps in knowledge appear within the framework of accepted concepts and mechanisms research is needed to fill those gaps. The models currently available can be used to scale estimates of transpiration from leaf to landscape levels and predict species responses to drought. The focus in tree water relations has shifted to examine the climatic thresholds at which drought, high temperatures and vapor pressure deficits cause mortality. Tree death may be caused by hydraulic collapse following irreversible cavitation or extremely low water potentials, but recent research indicates that the relative sensitivity of stomatal conductance and whole-plant hydraulic conductance plays a major role in determining plant responses to drought.