REPLYING TO M. Javaux, Y. Rothfuss, J. Vanderborght, H. Vereecken & N. Brüggemann Nature 536, 10.1038/nature18946 (2016)

In the accompanying Comment1, Javaux et al. correct a mistake in equation (2) of our work2; as they point out1, the mistake does not impact the central conclusion of our paper that ecohydrological separation is widespread2. However, equations (2) and (3) in ref. 2 calculate the source precipitation value of xylem water as the point where the xylem water evaporation line (EL) intersects the local meteoric water line (LMWL). In so doing, Javaux et al.1 note that the mistake affects our finding2 that “at 80% of the sites, the precipitation that supplies groundwater recharge and streamflow is different from the water that supplies parts of soil water recharge and plant transpiration”.

There are two key points in our response.

(1) We recognize the mistake now noted in equation (2) and thank Javaux et al.1 for this correction. These authors1 find that rainfall segregation could be observed at only 74% of the sites (as defined by the two criteria in ref. 1), and not 80% as we originally reported2.

(2) Our work2 presented evidence for ecohydrological separation based on a meta-analysis of isotopic dual liquid water isotope data (δ2H and δ18O) from 47 studies. This conclusion is supported by studies that analysed water vapour isotope data from the Tropospheric Emissions Spectrometer aboard NASA’s Aura satellite3 and by global differences between annual precipitation and groundwater isotope compositions4,5. These global-in-scale lines of evidence support earlier field evidence6,7 that ecohydrological separation (defined as plants using water of a character different to that of mobile water found in soils, groundwater and streamflow) is widespread, and is the rule rather than the exception. Ecohydrological separation was calculated using equation (1) in ref. 2.

It must be understood that equation (1) in ref. 2 is independent of the source precipitation analysis, which was calculated using equations (2) and (3) in that paper. Therefore, any issue with equation (2) in our paper, like the one raised by ref. 1, does not affect the ecohydrological separation conclusion.

We hope that this exchange will generate further interest in the use of stable O and H isotopes in plant water relation studies.