DELIMITING TREE TERRITORY FROM SOIL WATER BALANCE EQUATION: A CASE STUDY

Abstract

A usual problem linked to the apparent complexity of soil-plant interactions with the climatic forcing (as rainfall) is the need to compute the “active soil depth” (zr), available for water storage. Moreover, zr truncates the bottom depth of the control soil column, where drainage takes place, in the computation framework of a supply-demand-storage approach, in solving the soil water balance (SWB) equation at a site. Since the difficulty to estimate zr can lead to arbitrariness to set an operational value to it, our objective was to derive zr as the integral value of a unique soil layer of homogeneous properties. Our experimental scheme to solve the SWB equation involved a sample of ten young, 5 to 6 years-old, evergreen cork-oak (Quercus suber L.) trees, grown on a sandy loam soil (porosity is s), nearby Évora, Portugal. To achieve that goal, we have assembled several measurements techniques to assess the relevant water mass flux density terms, in the soil-plant-atmosphere continuum, as expressed per unit horizontal ground surface area (Ap), so justified by dimensional analysis. A steady-state IRGA (Li-1600M) apparatus was used to measure a discrete time series on instantaneous maximum diffusive conductance (gs) of sunlit leaf samples from four trees, under saturating light and soil water comfort, during spring season (in 2002). Further, gs was up-scaled to the corresponding daily foliage transpiration rate (Tr), which, in turn, was scaled up to plant root water uptake rate (U), using the derived sunlit leaf area index (Li) as the linear scale factor. In the same time period, the mean soil water variation rate (∆S), as induced by U, has been thermo-gravimetrically determined on collected soil core samples. Both ∆S and Tr was related to the spring season depletion phase of the ASW, lasted for a 30-day long period, the volumetric soil water () residence time (RT), under site conditions. Assumption of water mass conservation for the SWB problem in the rooted soil, as well as for whole-plant water capacitance, implies U matches ∆S, the depth-integrated value of  variation, throughout zr domain. Results showed zr can be estimated as a cumulative and directly proportional function of observed Tr and RT, without vertical discretization of the soil column. Model output showed an effective maximum rooting depth (or an equivalent lateral expansion) of 1754 mm (1.75 m), which is interpreted as the mean radius of an ellipsoidal zone of influence, in all directions; it was 1.32 times greater than the mean tree high. The product zr  s gives the “effective” soil depth, ca. 0.70 m, in this case study. Extension of the approach can be either determining U for pounded water land use system (e.g., rice cultivation) or estimating the plant density in irrigated orchards, enhancing water management.