G.B. Allison

The distribution of deuterium and 18O in dry soils: 1. Theory

Abstract A mathematical model is developed describing the shape of H 2 18 O and HDO depth profiles which result from evaporation of water from dry soil under quasi-steady state conditions. Typically, isotope concentrations rise from a minimum at the soil surface to a maximum a short distance beneath the surface, and then decrease approximately exponentially to constant concentrations at depth. The model predicts that for isothermal conditions, the slope of the relationship between 18 O and deuterium δ-values of samples of the soil water will be ∼30% lower for a dry soil than for a wet soil evaporating under the same conditions. It is concluded that low slopes should be indicative of soil water or groundwater recharged under arid or semi-arid conditions. Using the shape of the 18 O and deuterium profiles, three independent methods for estimating evaporation for dry soils are developed. When water loss occurs by both transpiration and evaporation, the slope of the 18 O-D relationship should be slightly lower than that for a site where water loss occurs by evaporation alone.

The use of natural tracers as indicators of soil-water movement in a temperate semi-arid region

Abstract In a semi-arid area of southern Australia a change in land use from Eucalyptus scrub to cropping with wheat is shown to have caused considerable change in the mechanism of the movement of soil water and the amount of deep drainage. Chloride concentrations of soil water have been used to show the mean annual amount of deep drainage increases from less than 0.1 to ∼ 3 mm yr. −1 following clearing of the native vegetation. The concentration of environmental tritium in soil water beneath the native vegetation is consistent with the hypothesis that some relatively recent water (post 1960) has penetrated to depths of at least 12m along channels occupied by living roots. Where the native vegetation has been cleared, no water which fell as rain since 1960 was found at depths greater than 2.5 m. 18 O and deuterium concentrations suggest that some water movement to the roots of the native vegetation is in the vapour phase.

The distribution of deuterium and 18O in dry soils 2. Experimental

Abstract This paper presents experimental data on the depth distribution of the stable isotopes 18 O and deuterium in soil water for columns of soil undergoing evaporation. Under steady-state and isothermal conditions in dry soil, isotope concentration rises from a low value at the soil surface to a maximum in the region where liquid transport of water dominates. The concentration then falls exponentially. We find agreement between observed isotope profiles and those calculated by theory developed in a previously published companion paper. Coefficients of determination for regressions of deuterium on 18 O delta values are high for all columns. Observed slopes ranged from 4.2 for saturated sand to 2.5 for sand having very low water content near the surface. This difference is explained in terms of increased path length for laminar diffusion of H 2 18 O and HDO in the vapour phase. This effect was verified by allowing water to evaporate through layers of dry porous material of varying thickness. In these experiments the slope varied between 3.9 and 2.7, showing reasonable agreement with that predicted by the model.

Recharge in karst and dune elements of a semi-arid landscape as indicated by natural isotopes and chloride

Abstract The rates and mechanisms of local recharge in a semi-arid environment have been investigated beneath two major landscape settings in the Murray Basin in South Australia; these were calcrete flats with sinkholes and sand dunes in adjacent landscape settings. Five minor settings were investigated and identified as undisturbed calcrete, primary (older) sinkholes, secondary (younger) sinkholes, sand dunes with native vegetation and sand dunes with introduced pasture. Data from a limited series of boreholes were interpreted to suggest that recharge varied from in excess of 100 mm yr.−1 for secondary sinkholes to less than 0.1 mm yr.−1 beneath sand dunes with native vegetation. The data suggest that recharge under sand dunes has increased by more than two orders of magnitude following clearing of the native vegetation ∼ 50 yr. ago. As ∼ 90% of an area of 10,000 km2 has been cleared, this clearing has ensured that inflow of saline groundwater to the adjacent River Murray will increase and, in view of the amount of salt stored in both the unsaturated and saturated zones, will continue for a very long time. At sites where the inferred recharge is lowest, profiles of the chloride concentration of soil water are consistent with the hypothesis that there is in excess of 50,000-yr. recharge stored in the thick unsaturated zone. Some evidence for changes in recharge rate over this time scale is seen.

The relationship between 18O and deuterium in water in sand columns undergoing evaporation

Abstract Experimental data show that the slope of the relationship between deuterium and 18 O δ-values can fall as low as 2 for soil water in unsaturated sand subjected to evaporation. This is considerably lower than the value of ∼5 obtained during evaporation of free water. The effect is explained in terms of increased thickness of the laminar layer through which evaporating water molecules escape. This work suggests that arid-zone groundwaters replenished by local recharge, should be characterised by low 18 O-D slopes.

Seasonal chemical and isotopic variations of soil CO2 at Trout Creek, Ontario

Apparatus for altering the frequency of a signal by values other than whole number division comprising the use of an adder and a register for consecutively and cumulatively adding an input number at a rate coincident with the frequency of a signal to be frequency divided and providing an output indicative of each occurrence of the cumulative sum exceeding a predetermined value.

Estimation of evaporation from the normally «dry» lake Frome in South Australia

Depth profiles of deuterium composition, expressed as delta-values, have been used to make point estimates of evaporation from the floor of the normally “dry” Lake Frome, a salt lake in northern South Australia. Values from five sites, varying between 90 and 230 mm yr.−1, have been used to obtain a relatively crude approximation for mean evaporation rate from the lake of 170 mm yr.−1. Under the assumption that this represents discharge of water from the regional groundwater system, the groundwater inflow into the lake is calculated as ∼ 5 × 108 m3 yr.−1. Two possible models which describe groundwater inflow into the lake are discussed.

The distribution of deuterium and 18O in dry soils: 3. Theory for non-isothermal water movement

A model is developed which predicts the shape of H218O and HDO depth profiles which result from evaporation of water from a soil under non-isothermal and quasi-steady-state conditions. The profiles have similar form to those which develop under isothermal conditions. However, when the soil at depth is cooler than the soil surface — the condition under which most evaporation takes place — a minimum develops in the isotope profile some distance beneath the evaporating front. This is due to the diffusive flux of isotopes in the vapour phase partially offsetting the convective isotope flux in the liquid phase. It is shown that for coarse-textured soils under isothermal conditions the diffusive vapour flux of isotopes may be the dominant mechanism in balancing convective effects in the liquid phase, even when no net movement of water vapour occurs. The theory is extended to enable partitioning of the water flux into liquid and vapour components under both isothermal and non-isothermal conditions.

Estimation of isotopic exchange parameters, using constant-feed pans

The use of constant-volume evaporation pans for estimating the isotopic composition of water evaporating from an open water body, δE, has been studied. Earlier work showed that pans in which water evaporates to dryness are not very useful for estimating δE because variations in meteorological conditions radically upset the approach to equilibrium. This problem is overcome by using constant-volume evaporation pans, and it is shown to be quite feasible to estimate “flux-weighted” values of δE which are appropriate for periods of several weeks or months. As well as having this important advantage over conventional pans, the concentration of the feed water can be modified to minimize the effect of any errors in the exchange parameter, m.

The movement of isotopes of water during evaporation from a bare soil surface

Stable isotopes of water (2H, 18O) at natural levels have been used to measure steady-state evaporation from soil surfaces in situations where other techniques would fail. In an endeavour to generalise the technique, we have investigated the movement of isotopes during nonsteady evaporation, especially during second-stage evaporation. Equations describing the isotope profiles are developed and it is found that the isotope concentration is a function of the Boltzmann variable (depth time−12). The equations are then solved under the assumption that the evaporation front is narrow. Seven packed sand columns were allowed to evaporate under controlled conditions and then sectioned for analysis of water content and isotopic composition. The results agree generally with the theory but there are difficulties with using isotopes to estimate cumulative evaporation.