Figure 9: Conceptualization of the water signals and seasonality patterns across the forest water cycle. During most of the year the xylem water signatures are more consistent with winter precipitation signatures.
Conclusion
Based on two years of stable water isotope measurements in precipitation, mobile and bulk soil waters, and beech and spruce xylem waters at our mixed forest site, we documented the seasonal signals and patterns of tree water uptake across the forest water cycle (Figure 9). We found that mobile and bulk soil waters exhibit distinct seasonal signals, with amplitudes that decrease with depth (Figure 2).
Recent precipitation was only dominant in a few xylem samples, collected predominantly in fall and spring (Figure 3). Recent precipitation made up a larger fraction of mobile soil waters than bulk soil waters (Figure 7); however, isotopic signals of bulk soil waters up to 40 cm depth also exhibit a seasonal cycle similar to precipitation (Figure 2). Mobile soil waters sampled at the same spot from lysimeters throughout the whole observation period were less variable compared to bulk soil waters (Figure 2). We found that xylem and bulk soil exhibited smaller differences than mobile soil water between dry and wet antecedent conditions. Mobile soil waters sampled during summer contained more summer precipitation when sampled following drier antecedent conditions (Figure 8).
Peak growing season xylem signatures match bulk soil signatures well, indicating that bulk soil waters are plausibly the major source of tree water uptake. Mixing calculations revealed that the isotope ratios in xylem water were a mixture of shallow (10 to 20 cm depth) and deep (40 to 80 cm depth) bulk soil waters for most sampling dates (Figure 4; Table 1). Beech trees predominantly sourced water from depths between 40 to 80 cm and spruce trees from 10 to 40 cm. Xylem waters exhibited a winter precipitation signature in both summer and winter, suggesting that trees at our site preferably use winter precipitation for transpiration (Figure 6; Figure 9). This holds also after potential uncertainties from δ2H depletion originating from cryogenic vacuum distillation or xylem water storage are accounted for.
In summary, beech and spruce forest trees at our site do not typically consume recent precipitation, but instead a mixture of bulk soil waters dominated by winter precipitation (Figure 9).