Diurnal variation in xylem water isotopic signature biases depth of root-water uptake estimates

Abstract

Total word count (ex. summary, references and legends): 6 932 words 25 of which: Summary (198 words); Introduction (698 words); Material and Methods (2 922 words); Results (1 138 words); Discussion (1 556 words) Conclusion (222 words); Acknowledgement (128 words) Number of color figures: 7 figures 30 Number of Tables: 2 tables Supplementary information: 7 (color Fig S1-S5 + Table S1-S2) 2 Summary  Stable water isotopes are a powerful and widely used tool to derive the depth of root water uptake (RWU) in lignified plants. Uniform xylem water isotopic signature (i-35 H2O-xyl) along the length of a lignified plant is a central assumption, which has never been properly evaluated. Here we studied the effects of diurnal variation in RWU, sap flow velocity and various other soil and plant parameters on i-H2O-xyl signature within a plant using a mechanistic plant hydraulic model. 40  Our model predicts significant variation in i-H2O-xyl along the full length of an individual plant arising from diurnal RWU fluctuations and vertical soil water heterogeneity. Moreover, significant differences in i-H2O-xyl emerge between individuals with different sap flow velocities. We corroborated our model predictions with field observations from French Guiana and northwestern China. Modelled i-H2O-45 xyl varied considerably along stem length ranging up to 18.3‰ in δ 2 H and 2.2‰ in δ 18 O, largely exceeding the range of measurement error.  Our results show clear violation of the fundamental assumption of uniform i-H2O-xyland occurrence of significant biases when using stable isotopes to assess RWU. As a solution, we propose to include monitoring of sap flow and soil water potential for 50 more robust RWU depth estimates.

Using isotopes to incorporate tree water storage and mixing dynamics into a distributed ecohydrologic modelling framework

“…First, we assumed that extracted stem water was representative of mobile tree‐stored water in order to maintain a simplistic approach to the numerical representation of tree‐stored water. As demonstrated by Zhao et al () and De Deurwaerder et al (), there exists the possibility of isotopic variations within stems. Low variation in observed δ XYLEM of both hemlock and beech at each location (Figures and ) suggested that this simplification was reasonable.…”

“…Rothfuss and Javaux (), Barbeta, Ogée, and Peñuelas (), Penna et al (), De Deurwaerder et al (), and Wang et al () highlighted conceptual limitations of the EMMA approach in that we must assume that measured soil water isotopic compositions are representative of available water sources at the time of RWU. Goldsmith et al () demonstrated that spatial and temporal variability in soil water can influence RWU depth estimates.…”

“…Studies investigating RWU depths have been motivated by a desire to understand the prevalence of ecohydrologic separation (Evaristo, McDonnell, & Clemens, 2017), climate adaptation strategies of plants (Barbeta & Peñuelas, 2017;Huo, Zhao, Gao, Wang, & Pan, 2018;Brinkmann, Eugster, Buchmann, & Kahmen, 2019;, RWU competition among native and invasive species (De Deurwaerder et al, 2018), optimal revegetation schemes for modified landscapes (Wang, Fu, Lu, Wang, & Zhang, 2019;Wang, Fu, Lu, & Zhang, 2017), and hydroclimatic extremes following anticipated tree species succession and to better resolve the functional traits of trees in land surface models (Matheny, Mirfenderesgi, & Bohrer, 2017). Rothfuss and Javaux (2017), Barbeta, Ogée, and Peñuelas (2018), Penna et al (2018), De Deurwaerder et al (2019, and Wang et al (2019) highlighted conceptual limitations of the EMMA approach in that we must assume that measured soil water isotopic compositions are representative of available water sources at the time of RWU. Goldsmith et al (2019) demonstrated that spatial and temporal variability in soil water can influence RWU depth estimates.…”

“…Wen et al () observed minimal separation between water within Z. mays and transpired waters during peak daily transpiration suggesting a well‐mixed storage. De Deurwaerder et al () observed stem water isotopic variations along the lengths of six deciduous tree species in French Guiana and northwestern China, implying that mobile tree‐stored water is not necessarily well mixed. We therefore propose that (a) complete mixing, that is, transpired water draws uniformly from all tree‐stored water independent of water age, and (2) piston flow, where transpired water is sourced from the oldest tree‐stored water, could each represent a working hypothesis for modelling the transport of tree‐stored water from root to leaf.…”

Estimates of tree root water uptake from soil moisture profile dynamics

“…Adding this to SF measurements gauging different roots (Lott et al, 1996) and analyses of stable isotope concentration in the xylem water (Rothfuss and Javaux, 2017) could avoid overly simplistic assumptions about soil water availability and mixing. Analyses with higher temporal resolution could also elucidate further details about diurnal variations in xylem water isotopic signatures (De Deurwaerder et al, 2019). Moreover, higher spatial coverage and resolution using hydrogeophysical, quantitative measurements like time-lapse ground penetrating radar (Allroggen et al, 2017;Jackisch et al, 2017) would enable further analyses of the active rhizosphere and its geometry.…”