Including tree water potential in plot-level transpiration modeling using the hydrodynamic approach

Above-ground water storage in trees plays a key role in regulating transpiration in forest canopies. Plants transpire water from the stem storage. As transpiration rates are higher than the maximal recharge rate from the soil through the roots, stem and branches, the above-ground storage becomes depleted and stomata close to restrict transpiration in response to the negative xylem water potential. These hydrodynamic limitations are known to control transpiration in forest ecosystems under both wet and dry conditions. Current land-surface models do not represent the above-ground storage in trees. These models impose water resource limitations on transpiration by directly linking stomatal conductance to soil moisture and soil water availability. As the intra-daily dynamics of soil moisture are slower and very different than the faster dynamics of water storage in the tree xylem, the current approach that do not incorporate tree-water storage leads to deviations from the observed dynamics of transpiration. As a result, land surface models produce characteristic intra-daily patterns of errors in simulations of latent heat flux. We propose a framework to resolve such tree hydrodynamics, which could be incorporated into hydrologic, land surface, and Earth System Models to replace the current empirical link between stomatal conductance and soil moisture. The FETCH2 model resolves the water flow and water potential in the tree stem and realistically links stomatal conductance to the water potential in the xylem, while water availability in the soil provides a bottom boundary condition for the hydrodynamic system. We use data from a large scale ecological disturbance experiment at a forest in Michigan to validate this approach. We compare the results of FETCH2 simulations to observations of sap flux and stem water storage made in the forest plots, and to land-surface model simulation results without the hydrodynamic module. Improvements to the simulation of stomatal conductance and transpiration in atmospheric, hydrologic, and Earth systems Models will propagate to connected variables such as soil moisture, the surface energy budget, and gross primary productivity. By incorporating the effects of forest canopy structure, tree water storage, and hydraulic strategy on stomatal conductance, our model developments may have a large impact on the resolved water cycles of many types and classes of models.