Constraints of using in-situ water vapour stable isotope analysis for estimating groundwater recharge in a beech forest

As temperatures rise, heat waves, droughts, and intense rainfall are expected to become more common and severe, posing significant risks to forest ecosystems. More frequent droughts make forests progressively vulnerable, leading to increased tree mortality particularly among drought-sensitive species like European beech (Fagus sylvatica S.). Understanding the interactions between forests and the water cycle is crucial to predict how forest ecosystems will respond to climate change and to develop adapted management strategies accordingly. By analysing the stable isotopes of water (δ²H, δ¹⁸O) which act as a natural fingerprint, we elucidate how beech trees cope with climate extremes and quantify water fluxes across the soil-plant-atmosphere continuum. Among these fluxes, groundwater recharge is essential for replenishing groundwater storage and sustaining stream baseflow. Here, we focus on estimating groundwater recharge under natural rainfall conditions, drought, and after extreme rainfall using HYDRUS-1D.

This study is conducted in a mature beech stand in the Rosalia forest located in the alpine forelands of Austria. The elevation at the study site is 650 m with an average slope of 16°. The mean annual precipitation is 790 mm, 60% of which falls between May to October, and the mean annual temperature is 8.2 °C. The soil is predominantly Cambisol, exhibits strong heterogeneity, and consists of 41% sand, 46% silt and 13% clay.

Climate change scenarios are simulated with rain‑out shelters (6x6 m) that induce drought stress in two trees and the surrounding soil. Sprinklers simulate extreme rainfall (75 mm per event) at two‑month intervals during the growing season, while two other trees serve as references under natural rainfall conditions. Soil water isotope profiles are collected via two complementary approaches: first, 100 cm soil cores are subdivided into 10 cm increments and analysed in the laboratory using the direct liquid-vapour equilibration method every three weeks. Second, since July 2025, in-situ soil water vapour is sampled within the rooting zone of one drought-treated and one reference tree at 10, 20, 30 and 60 cm, and analysed with an isotope ratio spectrometer (Picarro L2130-i) for daily measurements. These isotope data are supported by meteorological data including isotopic composition of precipitation, soil moisture, and matric potential.

Results showed that the soil exhibits strong heterogeneity in both isotopic composition and physical properties, with three to four soil horizons identified within the top 100 cm. Following irrigation, the isotope profile was largely replaced by the irrigation water isotope ratio within 100 cm, indicating preferential flow and rapid infiltration. We found strong temporal heterogeneity in soil water isotope profiles, and the isotopic profiles from in-situ vapour sampling and soil cores were only partly comparable, likely reflecting differences in isotopic composition of bulk water (core samples) and mobile water fractions (in-situ analysis), soil heterogeneity, and possibly method-specific biases. These discrepancies currently prevent robust estimates of groundwater recharge estimation with different approaches, underscoring the difficulty in applying these methods in strongly heterogeneous environments. Ongoing work includes system refinements and experimental redesign, alongside evaluation of more suitable methods to enable groundwater recharge quantification.