La Trobe Research: John Webb
Dryland salinity is one of the most important problems facing Australian agriculture today, as salinisation reduces the productivity of large areas of farmland. Developing management strategies relies on a complete knowledge of the processes that are causing the salinisation; although these are understood at a general level, they vary greatly at a local (paddock) scale.
The Environmental Geoscience group at La Trobe University has been studying dryland salinisation in western and central Victoria for over 8 years, in conjunction with the Glenelg-Hopkins and Wimmera Catchment Management Authorities, Primary Industries Research Victoria (Bendigo) and local landcare groups. The Australian Institute for Nuclear Science and Energy has provided funding for groundwater dating.
Our work has focussed on two aspects: groundwater chemical and isotopic studies, and the effects of landuse (vegetation) and climate on groundwater levels, through very detailed studies of individual sites and medium-long term modelling. There are 3 current and 2 completed PhD projects on these topics, along with 2 MSc and 15 Honours projects, and additional PhD projects are offered. The results have appeared in a number of publications and have been presented at numerous international conferences.
Chemical and isotopic studies
We use chemical and isotopic techniques to determine how groundwater flows through the aquifers and what is causing groundwater to come to the surface in particular locations and develop areas of salinisation. The results of these studies have been used to help manage the groundwater resources of the region.
Our studies demonstrate that soil processes, particularly plant uptake and interactions with clays, quickly change the chemical composition of infiltrating rain water to that of dilute seawater. Salinisation is predominantly due to evaporation where groundwater lies close to the surface, even if the groundwater at depth is fresh. Evaporation may cause salt to build up in low permeability clay-rich soils, but does not necessarily lead to salinisation if the salt lies below the pasture root zone, so productive farmland is frequently underlain by salty soils and saline groundwater.
On the basalt plains of western and central Victoria we have shown that many saline scalds and lakes are related to the geology: edges of basalt lava flows, and upwards groundwater flow from the buried river valley sediments beneath. Recharge to the groundwater in the basalts and underlying sediments occurs mostly through the volcanoes; the thick clayey soils of the plains allow only small amounts of saline recharge. However, the addition of this saline soil water causes the groundwater to progressively increase in salinity down the flow path. Overall, recharge is determined largely by soil texture rather than vegetation cover.
Effects of landuse (vegetation) and climate on groundwater levels
A very detailed study of a farm paddock using data loggers to record hourly groundwater fluctuations showed that drainage lines carrying water only intermittently act as the main recharge area. The drainage line also has very saline soils and groundwater because the watertable is shallow and there is insufficient flow to flush the salt. Thus replanting vegetation along ephemeral drainage lines will intercept runoff and reduce recharge and salinisation.
Tree plantations are believed to reduce recharge to groundwater, but preliminary detailed studies on a plantation, again using data loggers, have shown that there is little difference to nearby farmland. Preferential recharge may be occurring during heavy rainfall events down the tree roots.
Hydrological modelling using a monthly time-step has been carried out on a number of lakes in western and central Victoria, to determine the controls on lake level and salinity. This has shown that groundwater is rarely a significant component of the lake budget, and that saline baseflow is more important than evaporation in increasing salinity in water supply reservoirs. Climatic fluctuations are the primary factor in lake water and salt budgets, although land use changes have an impact in some instances. The models allow prediction of future trends under any likely climatic scenario.
This modelling has been extended to watertables in the basalt plains of western Victoria, by relating infiltration (from local climate data) to watertable fluctuations. Initial progress is promising, and the model is presently being refined by determining the best measure of infiltration and the time lag between infiltration events and watertable response. The results will then be tested against existing catchment models like CAT (Catchment Area Tool) and MIKE SHE, to see how different management strategies (tree planting, level of water usage) will impact on groundwater levels.