Peter J. Thorburn

Sources of water used by riparian Eucalyptus camaldulensis overlying highly saline groundwater

Water sources of Eucalyptus camaldulensis Dehn. trees were investigated on a semiarid floodplain in south-eastern Australia. The trees investigated ranged in distance from 0.5 to 40 m from a stream, with electrical conductivity 0.8 dSm−1, and grew over groundwater with electrical conductivity ranging from 30 to 50 dSm−1. The sources of water being used by the trees were investigated using the naturally occurring stable isotopes of water and measurements of soil water potential. Xylem water potential and leaf conductance were also examined to identify the trees response to using these sources of water. Trees at distances greater than about 15 m from the stream used no stream water. The trees used groundwater in summer and a combination of groundwater and rain-derived surface-soil water (0.05–0.15 m depth) in winter. In doing so they suffered water stress at electrical conductivities higher than approximately 40 dSm−1 (equivalent to approximately −1.4 MPa). Trees adjacent to the stream used stream water directly in summer, but may have used stream water from the soil profile in winter, after the stream had risen and recharged the soil water. E. camaldulensis appeared to be partially opportunistic in the sources of water they used.

Salt accumulation in semi-arid floodplain soils with implications for forest health

Dieback of native Eucalyptus largiflorens forests is an increasing problem on the floodplains of the lower River Murray, southern Australia. Salinisation of floodplain soils, as a result of the changed hydrological management of the River Murray, appears to be a primary cause of the dieback. Regulation of the River Murray has reduced the frequency of large flood events by a factor of approximately three and caused groundwater levels beneath floodplains to rise. The higher water tables have resulted in increased discharge of the naturally saline groundwater in the floodplains by evapotranspiration, and the decreased incidence of large floods has reduced floodwater recharge and hence leaching of salt from floodplain soils. Use of soil physical properties for a range of floodplain soils, combined with measurements of groundwater discharge from bare and vegetated sites, suggests that the time-scale for complete soil salinisation can, at worst, be less than 20 years. Moreover, salt accumulation at most sites will continue to occur as the present flooding regime (of which there is limited scope for improvement) appears incapable of providing the leaching required to counteract accumulation. The analyses carried out here suggest that the critical water table depth (below which groundwater discharge is balanced or exceeded by floodwater recharge) needs to be increased by 14 55% (the more clayey the soil, the larger the increase) to prevent salt accumulation. Failure to implement schemes which lower the water tables beneath the floodplain may, in the long term, cause serious damage to these important riparian forests.

Variations in stream water uptake by Eucalyptus camaldulensis with differing access to stream water

The stable isotopes 2H and 18O were used to determine the water sources of Eucalyptus camaldulensis at three sites with varying exposure to stream water, all underlain by moderately saline groundwater. Water uptake patterns were a function of the long-term availability of surface water. Trees with permanent access to a stream used some stream water at all times. However, water from soils or the water table commonly made up 50% of these trees water. Trees beside an ephemeral stream had access to the stream 40–50% of the time (depending on the level of the stream). No more than 30% of the water they used was stream water when it was available. However, stream water use did not vary greatly whether the trees had access to the stream for 2 weeks or 10 months prior to sampling. Trees at the third site only had access to surface water during a flood. These trees did not change their uptake patterns during 2 months inundation compared with dry times, so were not utilising the low-salinity flood water. Pre-dawn leaf water potentials and leaf 13C measurements showed that the trees with permanent access to the stream experienced lower water stress and had lower water use efficiencies than trees at the least frequently flooded site. The trees beside the ephemeral stream appeared to change their water use efficiency in response to the availability of surface water; it was similar to the perennial-stream trees when stream water was available and higher at other times. Despite causing water stress, uptake of soil water and groundwater would be advantageous to E. camaldulensis in this semi-arid area, as it would provide the trees with a supply of nutrients and a reliable source of water. E. camaldulensis at the study site may not be as vulnerable to changes in stream flow and water quality as previously thought.

Uptake of saline groundwater by plants: An analytical model for semi-arid and arid areas

An analytical model, based on unsaturated zone water and solute balances, was developed to describe the uptake of saline groundwater by plants in dry regions. It was assumed that: i. initially, the profile had low water and salt contents to some depth; ii. both water and solutes move upwards from the water table by piston flow due only to plant water extraction; iii. the uptake of water concentrates solutes in the soil solution until some threshold salinity is reached, above which plants can no longer extract water due to osmotic effects; iv. uptake of the groundwater does not affect the water table level; and v. uptake of groundwater is only limited by transmission of groundwater through the soil. Model predictions were compared with measurements of groundwater uptake made over 15 months at five sites in aEucalyptus forest in a semi-arid area, using independently measured model parameters. Depth and salinity of groundwater, and soil type varied greatly between sites. Predicted groundwater uptake rates were close to measured values, generally being within ∼ 0.1 mm day-1. Sensitivity analysis showed that groundwater depth and salinity were the main controls on uptake of groundwater, while soil properties appeared to have a lesser effect. The model showed that uptake of groundwater would result in complete salinisation of the soil profile within 4 to 30 yr at the sites studied, unless salts were leached from the soil by rainfall or flood waters. However, a relatively small amount of annual leaching may be sufficient to allow groundwater uptake to continue. Thus groundwaters, even when saline, may be important sources of water to plants in arid and semi-arid areas.

The Source of Water Transpired by Eucalyptus camaldulensis: Soil, Groundwater, or Streams?

Publisher Summary This chapter describes the relative importance of various possible sources of water available to river red gums or Eucalyptus camaldulensis. The technique used in this study monitored the stable isotope composition of water from tree sap and different possible water sources. However, a detailed analysis of isotope data and measurement of soil water status was used to provide insights into the water relations of Eucalyptus camaldulensis. The study not only addresses an important ecological problem but also provides an example of the power of isotopic methods for studying plant-water sources. Naturally occurring differences in the concentration of the stable isotopes D and 18 O were used to determine the sources of water transpired by E. camaldulensis. The comparison of tree sap and creek water δD and δ 18 O data showed that creek water is not the dominant source of water for trees flooded by creek water. The combination of the δD and δ 18 O data allowed application of a simple two end-member mixing model to estimate the proportion of groundwater and shallow soil water used by the trees.

EFFECT OF LAND DEVELOPMENT ON GROUNDWATER RECHARGE DETERMINED FROM NON-STEADY CHLORIDE PROFILES

Abstract The effect of clearing and subsequent crop and pasture growth on recharge to ground waters was investigated in three experimental catchments in the brigalow (Acacia harpophylla) lands of north-eastern Australia. Recharge was calculated from soil chloride data, using a simple transient solute mass balance model. Clearing had a substantial initial effect on groundwater recharge, with average recharge rates of 29 to 70 mm year−1 in two cleared catchments, compared with 7 mm year−1 in an uncleared catchment. These results were attributed to record high rains that fell while both cleared catchments were bare of vegetation, before crops or pastures were established. The effect was only short lived, however, with no significant recharge occurring in any of the three catchments during the period in which crops and pastures were fully established. This lack of recharge was contrary to the general belief that clearing and establishment of crops or pastures causes a sustained increase in groundwater recharge. The low recharge rates at this site were attributed to the slowly permeable soils and the climate of the study site, where potential evaporation exceeds average rainfall in all months, and to the water use characteristic of brigalow. Brigalow is shallow-rooted, and at this site generated lower soil water deficits than either crops or pastures. Clearing is unlikely to result in high water-tables in these soils under pastures or opportunity cropping systems under the average climatic conditions of the region. The simple transient solute mass balance model used to estimate recharge rates gave important and significant differences in recharge when compared with a more commonly used steady-state model. All recharge rates calculated with the steady-state model were ⩽ 1.8 mm year−1 (most

Comparison of diffuse discharge from shallow water tables in soils and salt flats

Abstract Diffuse (evaporative) discharge of ground water is of interest in the management of local or regional ground waters, and soil salinity. However, past studies show that discharge may vary greatly between soils in agricultural areas and salt flats for similar water table depths. Low discharge from salt flats has been previously attributed to the effect of salt crusts, yet possible soil hydrological reasons for those differences have not been examined. Steady-state hydraulic theory describing the relationship between discharge and water table depth is reviewed. The minimum water table depths required for the theory to be applied are defined in terms of soil parameters. Relationships between discharge and water table depth are then used to analyse the results of previous diffuse discharge studies. It is shown that discharge from both bare agricultural soils and salt flats is consistent with this theory. Unsaturated hydraulic conductivity of three salt flat soils, determined from measurements of discharge and soil matric suction, showed that low discharge flexus recorded from the sites were due to low soil permeability. The relationship between discharge flux and water table depth calculated for these sites also described discharge from other salt flats, implying that low hydraulic conductivity caused low discharge from these areas as well. The reasons for the low hydraulic conductivity of salt flat soils are not clear, and need to be investigated to determine if it is a general property of soils in these areas, or results from the high salinity levels.

The Use of Stable Isotopes of Water for Determining Sources of Water for Plant Transpiration

Over the last ten years, there has been a large increase in the number of vegetation studies that have incorporated measurements of the stable isotopic composition of water. There are many methods for measuring the amount of water being used by plants, but until recently it has been difficult to determine from where plants obtained their water. This has been particularly difficult where there is more than one available water source (e.g. where groundwater is shallow or streams are nearby).