Assessment of current and future salinity risk within the Queensland Murray-Darling Basin (QMDB) is essential to the protection of assets at both the local and Basin scale. This thesis explores aspects of salinity risk for the QMDB via using catchment salt balances and examines the level of complexity required in risk assessment.
The components of the salt mass balance and the balance itself (Export/Import ratio – E/I) are useful tools to encapsulate salinity processes within catchments. Salt inputs (atmospheric, anthropogenic, mineral weathering), salt stores (surface water, regolith, groundwater) and salt exports (streamflow, dust, deep drainage) have been evaluated for the QMDB in a level of detail never previously undertaken.
Rainfall salt accessions are more easily accounted than dust accessions, although the limited data available suggests the latter may be important in the very long-term. Rainfall salt inputs to the QMDB have been re-examined and a new generation of equations related to distance to coast have been developed. As with elsewhere in Australia, rainfall salt input declines exponentially with distance inland and is a major source of salts in the QMDB. Regolith/bedrock, groundwater, and streamflow ionic data suggest that mineral weathering is a relevant process creating salt inputs in some geologies in the QMDB, in particular eastern upland landscapes and to a lesser degree the Cretaceous geologies in the central and western QMDB. Further analysis of data is required however to achieve accurate quantification. Anthropogenic salt inputs have been calculated for the QMDB for the first time. Inputs via inter-basin water transfer, chlorination and fertiliser additions are negligible in the context of the QMDB, although inter-basin transfer is relevant in the Gowrie Creek catchment. Anthropogenic groundwater salt inputs are of a sufficient magnitude that they should be accounted for in catchment scale mass balance calculations – for example, in the Condamine catchment, groundwater salt inputs are potentially 4 times greater than atmospheric accessions. The expansion of the Coal Seam Gas (CSG) industry has the capacity to significantly increase non-cyclic groundwater salt accessions in the QMDB, in particular in the Condamine-Balonne catchments.
Salt stores in the QMDB have been evaluated using available data for surface water, groundwater and regolith materials. High regolith/groundwater salinity is common in the western two-thirds of the QMDB and large landscape salt stores are present. The conventional assumption that weathered zones in the landscape should be leached of salts is not borne out in either groundwater or regolith/bedrock data. Weathered zones may in fact have the highest salinity within a particular geological formation, although the relationship is not consistent. This likely reflects a change from historical (salt) transport unlimited conditions in the landscape during wetter climate phases, to the current transport limited conditions in a semi-arid climate i.e the accumulation of salts in weathered zones during drier climatic regimes post the period of weathering. Significant temporary surface water salt stores are held in the eastern half of the QMDB by virtue of large quantities of low salinity water being captured in dams, impoundments and other water bodies. While their role as a salt store is limited, these represent salt retained in the landscape that otherwise may have been exported via stream flow and can increase salt inputs to the landscape via increased recharge.
Streamflow is the primary salt export mechanism in the QMDB, although calculation of the salt load is dependent upon high quality data – a challenge in the episodic flow regime experienced in the region. Losses through groundwater recharge in clay dominated landscapes such as the QMDB are naturally small, although under post-clearing land uses such as irrigation they can be large and should be further investigated and considered in salt balance calculations. Dust may be a significant salt transport/loss mechanism out of the region, particularly over long time scales, but its magnitude remains very difficult to quantify. Fire-related and biomass salt export mechanisms have never previously been quantified, but example calculations suggest they are likely to only be minor components of the salt balance. The role of coal export as a salt loss mechanism requires further examination, as it is rapidly increasing in the QMDB and the coal is frequently washed with CSG water.
The salt balance within the QMDB is generally indicative of net accumulation of salts. This is particularly the case in larger, flatter, sedimentary rock catchments but less the case in small, eastern, hard-rock upland catchments. Many of these small catchments are net-exporters of salt although in a highly dynamic manner in response to climatic influence. It is suggested that the natural equilibrium for most catchments in the QMDB would have been a state of accumulation rather than an E/I ~1. The effects of clearing and land use change are more evident in terms of hydrologic and salinity response in smaller, steeper catchments than in the larger flatter catchments – the latter display trends that are only reflecting minor hydrologic change from loss of wooded cover. Consideration of relationships between E/I and catchment characteristics was undertaken for the first time, indicating complex interactions between geology, gradient, wooded cover and level of hydrologic disturbance. Comparison of long-term modelled E/I and annually calculated E/I suggests the latter is a more appropriate method for reporting in the QMDB as it accounts for episodic large salt export events driven by high rainfall inputs.
There has been considerable evolution of salinity risk assessment method in Queensland in the last decade, much of which has occurred in the QMDB during work by this author (although the work is not directly included in this thesis). Invariably, risk assessment approaches are limited by either the input data or by process understanding and the conceptualisation used. Detailed risk assessment methods used by the author have highlighted that such complex approaches do not always lead to a more ‘precise’ outcome. The spatial delineation of the conceptual models of SalCon (1997) remains the most efficient and practical approach that can be achieved with limited data. This makes an assumption that secondary salinity will express and places a focus on delineating the areas likely to be affected. It avoids the need for inclusion of estimates of salt store as there is invariably more than enough salt naturally occurring in landscapes to cause surface salinisation in appropriate conditions.
If a more quantitative approach to salinity risk assessment is required – for instance to predict future impacts in streams – the current lack of data for unsaturated and saturated zone properties across the region remains a significant constraint. A simplistic conceptualisation, focussed on first principles and identifying areas with combinations of risk factors, provides a reasonable estimate of areas likely to be affected. This conceptualisation is supported by available monitoring data and should provide a focus for targeted investment in data collection. The highest risk areas identified in the QMDB using this approach are irrigation areas (without any groundwater use) in the lower Moonie, lower Border Rivers and lower Balonne catchments where the unsaturated zone is relatively thin and bedrock is relatively impermeable.
This thesis and associated work has significantly progressed the catchment scale understanding of salinity fluxes within the QMDB and the overall salt balances within catchments. At the same time however, it highlights that the level of landscape process understanding is still very limited and considerable investment in data collection is required before more robust risk predictions can be made.