Untangling the Uncertainties in Plant Water Source Partitioning with Isotopes
Haoqi Gai,Peijun Shi,Zhi Li
DOI: https://doi.org/10.1029/2022wr033849
IF: 5.4
2023-01-01
Water Resources Research
Abstract:Plant water source partitioning suffers from multiple uncertainty sources including tracers, depleted xylem water deuterium, and mixing models, but the uncertainties have rarely been systematically evaluated. Taking apple trees of different ages as example, we combined four tracers of different characteristics (2H, 3H, 17O, and 18O), two xylem water deuterium bias correction methods (soil water excess and stem relative water content), and four mixing models (IsoSource, SIAR, MixSIR, and MixSIAR) to quantify the total uncertainty and the uncertainty of each component. For the total uncertainty, the contributions of the mixing models were the largest (37%), followed by those of tracers (28%) and interaction of tracers and models (27%), while the contributions of other sources were the smallest (8%). In particular, the xylem water deuterium bias correction methods had an uncertainty of 2%, implying its minor role in deviating plant water source partitioning. After evaluating the performance of the three components with four target functions, we recommend tracer combination of 2H18O with MixSIAR as the best framework for water source partitioning of apple trees on the Loess Plateau. The decomposed uncertainties and recommended methods provide technical support and promote understanding of variable results in plant water source partitioning. Plants may absorb water from multiple sources, including precipitation, soil water of different layers, streamwater, and groundwater. Identifying the plant water source is important for management of plant productivity and water resources. In this context, the isotope-based methods have been widely applied to partition plant water sources. However, uncertainties have been detected from tracers, depleted xylem water deuterium, and mixing models. Despite that the above uncertainty sources have been separately explored by different studies, and the contribution of each uncertainty component to the total uncertainty has rarely been investigated. To explore this knowledge gap, we employ four isotopes, two xylem water deuterium correction methods, and four mixing models to generate the possible isotope-correction method-mixing model combinations. We compare the uncertainty of each component in these combinations to determine the steps that should be paid more attention in plant water source partitioning. Further, we recommend the best isotope-correction method-mixing model combination for plant water source partitioning. Although the recommended method may be limited to the regions with climate and vegetation similar to this study, this study presents a valuable technical insight into the evaluation and selection of methods for plant water source partitioning. We decompose the uncertainties in isotope-based plant water source partitioning, and analyze dominance of each uncertainty componentMixing models and tracers have dominant but deuterium bias correction has negligible effect on partitioned plant water sourcesWe propose a framework on evaluating and selecting the most suitable methods for plant water source partitioning