SummaryPlants rely primarily on rainfall infiltrating their root zones -a supply that is inherently variable, and fluctuations are predicted to increase on most of the Earth's surface. Yet, interrelationships between water availability and plant use on short timescales are difficult to quantify and remain poorly understood.To overcome previous methodological limitations, we coupled high-resolution in situ observations of stable isotopes in soil and transpiration water. We applied the approach along with Bayesian mixing modeling to track the fate of 2 H-labeled rain pulses following drought through soil and plants of deciduous tree ecosystems. We resolve how rainwater infiltrates the root zones in a nonequilibrium process and show that tree species differ in their ability to quickly acquire the newly available source. Sessile oak (Quercus petraea) adjusted root uptake to vertical water availability patterns under drought, but readjustment toward the rewetted topsoil was delayed. By contrast, European beech (Fagus sylvatica) readily utilized water from all soil depths independent of water depletion, enabling faster uptake of rainwater.Our results demonstrate that species-specific plasticity and responses to water supply fluctuations on short timescales can now be identified and must be considered to predict vegetation functional dynamics and water cycling under current and future climatic conditions.

Field studies analyzing the stable isotope composition of xylem water are providing important information on ecosystem water relations. However, the capacity of stable isotopes to characterize the functioning of plants in their environment has not been fully explored because of methodological constraints on the extent and resolution at which samples could be collected and analysed. Here, we introduce an in situ method offering the potential to continuously monitor the stable isotope composition of tree xylem water via its vapour phase using a commercial laser-based isotope analyser and compact microporous probes installed into the xylem. Our technique enables efficient high-frequency measurement with intervals of only a few minutes per sample while eliminating the need for costly and cumbersome destructive collection of plant material and laboratory-based processing. We present field observations of xylem water hydrogen and oxygen isotope compositions obtained over several days including a labelled irrigation event and compare them against results from concurrent destructive sampling with cryogenic distillation and mass spectrometric analysis. The data demonstrate that temporal changes as well as spatial patterns of integration in xylem water isotope composition can be resolved through direct measurement. The new technique can therefore present a valuable tool to study the hydraulic architecture and water utilization of trees.

Transit times through hydrologic systems vary in time, but the nature of that variability is not well understood. Transit times variability was investigated in a 1 m3 sloping lysimeter, representing a simplified model of a hillslope receiving periodic rainfall events for 28 days. Tracer tests were conducted using an experimental protocol that allows time‐variable transit time distributions (TTDs) to be calculated from data. Observed TTDs varied with the storage state of the system, and the history of inflows and outflows. We propose that the observed time variability of the TTDs can be decomposed into two parts: “internal” variability associated with changes in the arrangement of, and partitioning between, flow pathways; and “external” variability driven by fluctuations in the flow rate along all flow pathways. These concepts can be defined quantitatively in terms of rank StorAge Selection (rSAS) functions, which is a theory describing lumped transport dynamics. Internal variability is associated with temporal variability in the rSAS function, while external is not. The rSAS function variability was characterized by an “inverse storage effect,” whereby younger water is released in greater proportion under wetter conditions than drier. We hypothesize that this effect is caused by the rapid mobilization of water in the unsaturated zone by the rising water table. Common approximations used to model transport dynamics that neglect internal variability were unable to reproduce the observed breakthrough curves accurately. This suggests that internal variability can play an important role in hydrologic transport dynamics, with implications for field data interpretation and modeling.

In this commentary, we summarize and build upon discussions that emerged during the workshop "Isotopebased studies of water partitioning and plant-soil interactions in forested and agricultural environments" held in San Casciano in Val di Pesa, Italy, in September 2017. Quantifying and understanding how water cycles through the Earth's critical zone is important to provide society and policymakers with the scientific background to manage water resources sustainably, especially considering the ever-increasing worldwide concern about water scarcity. Stable isotopes of hydrogen and oxygen in water have proven to be a powerful tool for tracking water fluxes in the critical zone. However, both mechanistic complexities (e.g. mixing and fractionation processes, heterogeneity of natural systems) and methodological issues (e.g. lack of standard protocols to sample specific compartments, such as soil water and xylem water) limit the application of stable water isotopes in critical-zone science. In this commentary, we examine some Published by Copernicus Publications on behalf of the European Geosciences Union. 6400 D. Penna et al.: Tracing terrestrial ecosystem water fluxes using stable isotopes of the opportunities and critical challenges of isotope-based ecohydrological applications and outline new perspectives focused on interdisciplinary research opportunities for this important tool in water and environmental science.

Abstract. Determining the soil hydraulic properties is a prerequisite to physically model transient water flow and solute transport in the vadose zone. Estimating these properties by inverse modelling techniques has become more common within the last 2 decades. While these inverse approaches usually fit simulations to hydrometric data, we expanded the methodology by using independent information about the stable isotope composition of the soil pore water depth profile as a single or additional optimization target. To demonstrate the potential and limits of this approach, we compared the results of three inverse modelling strategies where the fitting targets were (a) pore water isotope concentrations, (b) a combination of pore water isotope concentrations and soil moisture time series, and (c) a two-step approach using first soil moisture data to determine water flow parameters and then the pore water stable isotope concentrations to estimate the solute transport parameters. The analyses were conducted at three study sites with different soil properties and vegetation. The transient unsaturated water flow was simulated by solving the Richards equation numerically with the finiteelement code of HYDRUS-1D. The transport of deuterium was simulated with the advection-dispersion equation, and a modified version of HYDRUS was used, allowing deuterium loss during evaporation. The Mualem-van Genuchten and the longitudinal dispersivity parameters were determined for two major soil horizons at each site. The results show that approach (a), using only the pore water isotope content, cannot substitute hydrometric information to derive parameter sets that reflect the observed soil moisture dynamics but gives comparable results when the parameter space is constrained by pedotransfer functions. Approaches (b) and (c), using both the isotope profiles and the soil moisture time series, resulted in good simulation results with regard to the Kling-Gupta efficiency and good parameter identifiability. However, approach (b) has the advantage that it considers the isotope data not only for the solute transport parameters but also for water flow and root water uptake, and thus increases parameter realism. Approaches (b) and (c) both outcompeted simulations run with parameters derived from pedotransfer functions, which did not result in an acceptable representation of the soil moisture dynamics and pore water stable isotope composition. Overall, parameters based on this new approach that includes isotope data lead to similar model performances regarding the water balance and soil moisture dynamics and better parameter identifiability than the conventional inverse model approaches limited to hydrometric fitting targets. If only data from isotope profiles in combination with textural information is available, the results are still satisfactory. This method has the additional advantage that it will not only allow us to estimate water balance and response times but also site-specific time variant transit times or solute breakthrough within t...

Abstract. Stable isotope signatures provide an integral fingerprint of origin, flow paths, transport processes, and residence times of water in the environment. However, the full potential of stable isotopes to quantitatively characterize subsurface water dynamics is yet unfolded due to the difficulty in obtaining extensive, detailed, and repeated measurements of pore water in the unsaturated and saturated zone. This paper presents a functional and cost-efficient system for nondestructive continual in situ monitoring of pore water stable isotope signatures with high resolution. Automatic controllable valve arrays are used to continuously extract diluted water vapor in soil air via a branching network of small microporous probes into a commercial laser-based isotope analyzer. Normalized liquid-phase isotope signatures are then obtained based on a specific on-site calibration approach along with basic corrections for instrument bias and temperature dependent isotopic fractionation. The system was applied to sample depth profiles on three experimental plots with varied vegetation cover in southwest Germany. Two methods (i.e., based on advective versus diffusive vapor extraction) and two modes of sampling (i.e., using multiple permanently installed probes versus a single repeatedly inserted probe) were tested and compared. The results show that the isotope distribution along natural profiles could be resolved with sufficiently high accuracy and precision at sampling intervals of less than four minutes. The presented in situ approaches may thereby be used interchangeably with each other and with concurrent laboratory-based direct equilibration measurements of destructively collected samples. It is thus found that the introduced sampling techniques provide powerful tools towards a detailed quantitative understanding of dynamic and heterogeneous shallow subsurface and vadose zone processes.

Abstract. Stable water isotopes are widely used in ecohydrology to trace the transport, storage, and mixing of water on its journey through landscapes and ecosystems. Evaporation leaves a characteristic signature on the isotopic composition of the water that is left behind, such that in dual-isotope space, evaporated waters plot below the local meteoric water line (LMWL) that characterizes precipitation. Soil and xylem water samples can often plot below the LMWL as well, suggesting that they have also been influenced by evaporation. These soil and xylem water samples frequently plot along linear trends in dual-isotope space. These trend lines are often termed “evaporation lines” and their intersection with the LMWL is often interpreted as the isotopic composition of the precipitation source water. Here we use numerical experiments based on established isotope fractionation theory to show that these trend lines are often by-products of the seasonality in evaporative fractionation and in the isotopic composition of precipitation. Thus, they are often not true evaporation lines, and, if interpreted as such, can yield highly biased estimates of the isotopic composition of the source water.

Distributions of water transit times (TTDs), and related storage‐selection (SAS) distributions, are spatially integrated metrics of hydrological transport within landscapes. Recent works confirm that the form of TTDs and SAS distributions should be considered time variant—possibly depending, in predictable ways, on the dynamic storage of water within the landscape. We report on a 28 day periodic‐steady‐state‐tracer experiment performed on a model hillslope contained within a 1 m3 sloping lysimeter. Using experimental data, we calibrate physically based, spatially distributed flow and transport models, and use the calibrated models to generate time‐variable SAS distributions, which are subsequently compared to those directly observed from the actual experiment. The objective is to use the spatially distributed estimates of storage and flux from the model to characterize how temporal variation in water storage influences temporal variation in flow path configurations, and resulting SAS distributions. The simulated SAS distributions mimicked well the shape of observed distributions, once the model domain reflected the spatial heterogeneity of the lysimeter soil. The spatially distributed flux vectors illustrate how the magnitude and directionality of water flux changes as the water table surface rises and falls, yielding greater contributions of younger water when the water table surface rises nearer to the soil surface. The illustrated mechanism is compliant with conclusions drawn from other recent studies and supports the notion of an inverse‐storage effect, whereby the probability of younger water exiting the system increases with storage. This mechanism may be prevalent in hillslopes and headwater catchments where discharge dynamics are controlled by vertical fluctuations in the water table surface of an unconfined aquifer.