Abstract: As one of the major types of groundwater dependent ecosystems, groundwater dependent terrestrial vegetation (GDTV) has a variety of ecological service functions.Protection of GDTV is an important criterion in sustainable water resources management, particularly when it is being threatened globally by degradation due to the over-exploitation of groundwater and surface water.Identifying GDTV is the first requisite step to managing and protecting it.While GDTV has been identified at large scales using the White method, water balance method, and satellite-based approaches, only stable isotope techniques can provide direct means for identifying GDTV.The stable isotopes approach can be further divided into direct inference approach and mixing model approach.Though only qualitatively identifying GDTV, the direct inference approach is still more prevalent than the mixing model approach.In recent years, with the application of Bayesian mixing models, progress has been made in quantifying the dependency of GDTV on groundwater by using mixing model approach.However, many conceptual and methodological challenges remain.The first one is how representative an individual tree/plant studies are for larger-scale systems, given the spatial and temporal heterogeneity of groundwater use pattern by GDTV, which also makes it difficult to applying our observational results at small-scale to understand catchment-or landscape-scale phenomena.Secondly, it has been found that many plants might use groundwater in an indirect way, which distorts the identifying of GDTV.The third main challenge is how the isotopic fractionation at soil-root interface and the spatial-temporal variation in xylem isotopic signatures within a plant can inform endmember determination.Furthermore, obtaining representative groundwater and soil water samples is also challenging.In the coming years, efforts should be made towards:(1) developing novel methods for in situ, online, and continuous isotopic measurements in xylem water, to help to gather isotope data at higher temporal and spatial resolutions; (2) conducting extensive isotopic labelling experiments with known boundary conditions, to facilitate a more detailed characterization of the flow and fractionation of isotopologues in the groundwater-soil-vegetation system; and (3) developing approaches coupling physically based ecohydrological models and isotopic observation, to provide root water uptake profiles with high spatial resolution andlower uncertainty.