Isotopic composition of bare soil evaporated water vapor. Part I: RUBIC IV experimental setup and results

https://doi.org/10.1016/j.jhydrol.2009.01.034Get rights and content

Summary

Stable water isotopes such as oxygen 18, are natural tracers of water movement within the soil–vegetation–atmosphere continuum. They provide useful information for a better understanding of evaporation and water vapor transport within soils. This paper presents a novel controlled experimental setup. It is dedicated to detailed measurements of the evaporation fluxes from bare soil columns, as well as to the corresponding isotopic composition of the water vapor, under non-steady state conditions. The experiment allowed an accurate determination of these quantities. The formulae encountered in the literature were used to estimate the isotopic composition of the evaporated water vapor. None of them was able to correctly reproduce the measured isotopic composition of water. The data were then used to estimate the value of the isotopic composition of the soil liquid water, which should be used to get the right results for the isotopic composition of the evaporated water vapor. Results suggest that, when liquid transfer is dominant within the soil, the isotopic composition of evaporation was controlled by the isotopic composition of the liquid water within very thin soil surface layers. When there is a peak in the isotopic profile, i.e. when water vapor is dominant close to the surface, the isotopic composition of the evaporated water seems to be governed by the isotopic composition of the soil liquid water at the peak.

The data were also used to estimate the kinetic fractionation factor. The results suggest that the latter is not constant in time. The values seem to depend on the shape of the isotopic profile. In both cases, the uncertainty on the results is very large. The estimation of the kinetic fractionation factor is studied more in details using the modeling results presented in Part II of a companion paper where the data set is modeled using the SiSPAT_Isotope model.

Introduction

Evaporation from soils and transpiration by vegetation represent the major rainfall-recycling source over continents (Chahine, 1992, Parlange and Katul, 1992, Costanza et al., 1997, Zangvil et al., 2004). Consequently, a correct assessment of potential impacts of water management practices, land use and/or climate change on water resources relies on an accurate representation of evapotranspiration within atmosphere, hydrological or vegetation models. Soil Vegetation Atmosphere Transfer (SVAT) models represent the complex interactions between the atmosphere, the soil and the biosphere. Most of them provide separate estimates of soil evaporation, interception by the canopy and transpiration by plants. However, few data (relying mainly on sap flow and micro-lysimeters measurements) are currently available to validate that partition. Stable water isotopes are natural tracers of water movement. They can provide useful information to quantify and understand this partition (Yakir and da Silveira Lobo Sternberg, 2000, Yepez et al., 2003, Williams et al., 2004). The isotopic composition of water within soils is known to be modified under soil evaporation (i.e. Barnes and Allison, 1983, Barnes and Allison, 1984), whereas no fractionation of isotopic forms of either oxygen or hydrogen occurs during root extraction (Zimmermann et al., 1967, Walker and Richardson, 1991, Bariac et al., 1994). The isotopic composition of evaporated and transpired water vapor is therefore expected to be different. In the field however, they cannot be measured separately as they are instantaneously mixed with the ambient air water vapor. It is therefore necessary to estimate them from measurements of the isotopic composition of liquid water in soils, leaves and stems (e.g. Yakir and da Silveira Lobo Sternberg, 2000, Yepez et al., 2003, Williams et al., 2004) with formulae established for water bodies or oceans such as the Craig and Gordon (1965) model (Yakir and da Silveira Lobo Sternberg, 2000). Their use for deriving the isotopic composition of soil evaporated water vapor is particularly critical, as soils are generally unsaturated and the evaporation front moves below the surface as the soil dries out. To study the mechanisms controlling the isotopic composition of evaporating soils, we have developed a physically based model, called SiSPAT_Isotope (Braud et al., 2005a) for bare soil, representing the full interactions between the atmosphere, the soil and stable isotope species. The model was evaluated against two sets of laboratory data (Braud et al., 2005b). The data were composed of soil columns which were let evaporating freely in the atmosphere. The results of this first study showed that lots of uncertainty in the modeling and interpretation of stable isotope composition of water in terms of evaporation were related to a lack of control of the experimental conditions, especially of the atmospheric relative humidity and to a lack of knowledge of the kinetic fractionation factor for unsaturated soils. In this paper, we present a novel controlled experiment dedicated to the measurement of the evaporation flux from bare soil columns, as well as its isotopic composition. The experimental setup allowed a precise determination of the atmospheric conditions, of the evaporation fluxes and of their isotopic composition. In addition, the data provide an evaluation of the formulae traditionally used for deriving the isotopic composition of bare soil evaporated water vapor, and especially of the relevance of the kinetic fractionation factor values proposed in the literature for free water bodies (Merlivat and Jouzel, 1979, Cappa et al., 2003). We also propose estimations of this kinetic fractionation factor and of its associated standard error with the experimental data. Data interpretation was found to be very sensitive to the value of the isotopic composition of the soil surface liquid water. It raises questions about the relevant sampling depth which is required to properly estimate the isotopic composition of evaporated water vapor. This point is studied more in details in Part II of a companion paper (Braud et al., 2009) which presents the modeling of these experimental results using the SiSPAT_Isotope model.

Section snippets

Experimental setup

The RUBIC IV reactor (Fig. 1, Fig. 2) was developed in order to determine the isotopic composition of the water vapor released by an evaporating soil and to monitor its time evolution as long as the soil was drying. The leak tight experimental setup allowed to inject directly a gas flow of dry air simultaneously over six evaporating soil columns and to continuously capture all the water vapor released by evaporation by cryoscopic trapping. Therefore, the only water vapor source was that of the

Evaporation flux

Fig. 6 shows two examples (for columns 1 and 4) of the comparison between the cumulative and instantaneous evaporation flux using the four (three) methods described in “Calculation of the evaporation flux and correction of the reactor dynamics” Table 5 provides the values of the cumulative evaporation for the six columns and all the methods. It shows that, apart from the ΔS method, the agreement between the methods is very satisfying. Differences on cumulative evaporation are less than 5 mm

Conclusions

In Part I of this paper, we have presented a novel experiment which allows to measure simultaneously the evaporation flux and the isotopic composition of the evaporated water under controlled conditions for bare soil. We compared the measured isotopic composition of the evaporated water with traditional estimates. The results show that, using experimental data, none of the kinetic fractionation factor values encountered in the literature was able to give results in agreement with the measured

Acknowledgement

The “Programme National de Recherche en Hydrologie” of the French “ECosphère COntinentale” program is acknowledged for providing the financial support of the experiment.

References (26)

  • E.A. Yepez et al.

    Partitioning overstory and understory evapotranspiration in a semiarid savanna woodland from the isotopic composition of water vapor

    Agric. Forest Meteorol.

    (2003)
  • C.D. Cappa et al.

    Isotopic fractionation of water during evaporation

    J. Geophys. Res.

    (2003)
  • M.T. Chahine

    The hydrological cycle and its influence on climate

    Nature

    (1992)
  • Cited by (54)

    • Isotope signature of maize stem and leaf and investigation of transpiration and water transport

      2021, Agricultural Water Management
      Citation Excerpt :

      Diurnal variations of the isotope compositions of leaf (δleaf), stem (δstem) and soil (δsoil) water show that δstem and δsoil had a similar distribution and variation both in sunny and cloudy day (Fig. 2a). Soil water isotope composition was a little higher than maize stem water, because of the strong evaporation of shallow soil water causing isotopic fractionation (Allison and Hughes, 1983; Braud et al., 2009). However, the δstem showed a crest value at noon of sunny day which higher than the soil water, indicating that isotopes fractionation occurred in processes of plant water absorption in study site.

    • Rainwater use by cotton under subsurface drip and center pivot irrigation

      2019, Agricultural Water Management
      Citation Excerpt :

      The measured isotopic signature of soil water in the profile showed that a stable soil water isotopic signature had developed on DOY 195 in the dryland field (Fig. 3). The soil profile showed a well-defined fractionation peak at 40 mm that has been previously observed at this location (Goebel et al., 2015) and is well described by others (Allison, 1982; Braud et al., 2009). The peak of enrichment prior to the rain event on DOY 196 was 5.5 δ18O (‰) at 40 mm with depletion occurring with depth to 160 mm and an isotopic signature of –4.0 δ18O (‰).

    View all citing articles on Scopus
    View full text