An assessment of contemporary variations of the Bradfield Scheme. A technical report to the National Water Grid Authority from the Bradfield Scheme Assessment

Petheram, C., Read, A., Hughes, J., Marvanek, S., Stokes, C., Kim, S., Philip, S., Peake, A., Podger, G., Devlin, K., Hayward, J., Bartley, R., Vanderbyl, T., Wilson, P., Pena Arancibia, J., Stratford, D., Watson, I., Austin, J., Yang, A., Barber, M., Ibrahimi, T., Rogers, L., Kuhnert, P., Wang, B., Potter, N., Baynes, F., Ng, S., Cousins, A., Jarvis, D., and Chilcott, C. (2022) An assessment of contemporary variations of the Bradfield Scheme. A technical report to the National Water Grid Authority from the Bradfield Scheme Assessment. Report. CSIRO, Sandy Bay, Hobart, Australia.

[img] PDF (Published Version) - Published Version
Restricted to Repository staff only



In 1938 Dr John Bradfield, an eminent engineer at the time, proposed an ambitious scheme to divert water via a series of dams and tunnels from the east-draining Tully, Herbert and Burdekin rivers on the north-east Queensland coast to the westerly draining semi-arid Flinders River and then to the arid internally draining Thomson River, which flows into Kati Thanda–Lake Eyre. Eighty-three years after first being proposed, the Bradfield Scheme and its variants still arise as part of the national discourse on drought and water security. Interest in ‘Bradfield concepts’ rise especially in times of drought and are generally promoted as means to stimulate ‘nation building’ through creating significant and enduring regional economic development opportunities in water supply and irrigation. Instead of diverting water to western Queensland, many contemporary variations of Bradfield’s scheme propose diverting water to the Murray–Darling Basin (MDB) where there is already an established irrigation industry and supporting infrastructure and demand for water. The Australian Government through the National Water Grid Authority commissioned CSIRO to undertake an assessment of contemporary variations to Bradfield’s scheme. This Assessment found that a partially solar photovoltaic (PV) powered pumped pipeline from the upper Tully catchment to the upper Herbert catchment and a ~62-m high dam and gravity diversion tunnel from the upper Herbert River to the upper Burdekin catchment, combined with runoff from the upper Burdekin catchment could generate a combined mean annual inflow of 2644 GL to a potential ~98-m high dam at Hell’s Gates on the upper Burdekin River, while ensuring sufficient water was released down the Tully, Herbert and Burdekin rivers to meet the needs of existing downstream entitlement holders. A 152-m (500 foot) dam, as proposed by Bradfield (1942) and some contemporary commentators, would never fill because the net evaporation from the reservoir surface would exceed the long-term inflows. After releasing water to meet the needs of downstream entitlement holders a potential ~98-m high Hell’s Gates dam could potentially release 2280 GL in 75% of years into a water supply channel with an offtake ~45 m above the base of the dam. This offtake height would enable a 1600-km gravity channel with a deep cutting or a slightly longer gravity channel with a 43-m re-lift station to convey water to St George on the Condamine-Balonne River in the northern MDB, the first major irrigation area along the potential channel alignment. Taking into consideration the 28% annual flow loss in the adopted channel configuration to St George, a pumped pipeline from the potential Hell’s Gates dam to St George had a levelised cost (i.e. the annualised cost divided by the mean annual diversion) greater than three times that of the channel configuration. There is little scope to enhance channel diversions between the potential Hell’s Gates dam and the northern MDB by capturing additional water en route. Major drainage lines that intersect the potential channel alignment have unfavourable topography for potential off-line channel storages, and resulting dams would be low yielding and costly. There is also limited potential to generate hydro-electric power along the backbone infrastructure water supply line. Opportunities to supply water to other industries along the water supply channel are also limited. The potential channel alignment traverses the most resource poor parts of Queensland, largely due to the extensive sedimentary cover. Furthermore, relative to agriculture, mining requires relatively little water and does not typically require high-quality water. Mining companies usually have sufficient resources to be self-sufficient in terms of their water requirements. No regional centres with water security issues are located near the potential channel alignment, and therefore the contemporary Bradfield Scheme offers little benefit for improving the water security of regional centres in Queensland. The optimal backbone infrastructure configuration (i.e. dams, pipes, tunnels and channels along the main water supply line) to St George is estimated to cost between $15 billion and $30 billion (assuming favourable geological conditions) with an annual cost of between $130 million and $255 million (including operation and maintenance, annual pumping costs and net revenue). After taking 7 to 10 years for approvals it is estimated that the backbone infrastructure would take a minimum of 12 years to construct. For a mix of ‘high’ priority (100% reliability) and ‘medium’ priority (75% annual time reliability) water, a mean of ~1270 GL of water could be diverted to St George after losses, which is equivalent to 25% of the average annual volume of water used for irrigation in the MDB between 2015 and 2019. It should be noted, however, in the Bradfield Scheme source catchments no releases were made to mitigate impacts to downstream water-dependent ecosystems or meet environmental flow objectives stipulated in the state government water plans. Releasing water for this purpose would reduce the volume of water that could be diverted. Adjacent and downstream of the existing Beardmore Dam (capacity 81.7 GL) near St George there is approximately 90,000 ha of land already developed for irrigated cotton and other broadacre crops and 600 ha of land already developed for irrigated horticulture. However, collectively, irrigators along the Condamine-Balonne River can only extract their full entitlement in about 40% of years, and this area of developed land is only fully irrigated in very wet years when ‘flood harvesting’ is possible. In drier years large extraction shortfalls occur because there are relatively few large dams and weirs to regulate the highly variable flows in the river, and the northern MDB more generally. Although high-value horticulture is approximately two to nine times more profitable per unit of irrigation applied than cotton, which is the most profitable broadacre/industrial crop, large-scale expansion of horticulture in the northern MDB is limited by the reliability of water supply, and as discussed later, ultimately, markets. South of St George along the Condamine-Balonne River and the nearby Moonie River there is more than 780,000 ha of soil potentially suitable for irrigated agriculture, of which about 150,000 ha are sandy and loamy soils potentially suited for horticultural crops. It was found that under the ‘optimal’ Bradfield Scheme backbone infrastructure configuration, which included a 43-m re-lift pump station and a large terminal storage, sufficient water could be delivered to increase the reliability of fully irrigating the existing 90,000 ha of cotton and broadacre cropping in 75% of years, as well as fully irrigating an additional 80,000 ha of new cotton in 75% of years and about 30,000 ha of new high-value horticulture in 100% of years. Thirty thousand hectares of citrus, the most profitable crop per megalitre of water consumed that could be grown in the St George region, was adopted for this analysis to represent an optimistic economic ceiling for horticultural production. Assuming the total cost of backbone and reticulation infrastructure to be $21 billion, under a set of extremely optimistic assumptions used to estimate an unattainable upper bound of financial performance, a discounted cash flow analysis over the lifetime of the scheme (100 years) showed water would have to be charged at $2310 for each megalitre supplied to cover the costs of the scheme. Net farm revenue was sufficient to afford no more than $580 per megalitre of water supplied, which would only cover about a quarter of the scheme’s costs. The cost of diversion infrastructure alone would add a premium of $1920 to the cost of each megalitre of water used to irrigate crops, making water cost about six times what it would without the diversion infrastructure. The inclusion of renewable energy and pumped pipelines made a very small change to the overall economics of the scheme, mainly because the cost of the water storage and particularly the main water supply channel from potential Hell’s Gates dam to the northern MDB dwarf all other costs. However, this analysis is highly optimistic. Infrastructure that involves substantial subsurface excavation, such as dams, tunnels and channels, have long construction periods and are particularly susceptible to large cost overruns. Australian and international studies of large-dam and mega-dam projects report a mean cost overrun of 120 and 100% respectively. Further, market projections estimate horticultural growth in the St George region would be unlikely to exceed 13,000 ha by 2050, even if unconstrained by the availability of water. Allowing for a modest combination of risks (including a 20% infrastructure cost overrun and slower more realistic expansion of horticulture) lowers the proportion of the revised scheme’s costs for which irrigators could pay to 8%. The Bradfield Scheme storage and diversion infrastructure offers minor financial benefit in mitigating flooding in the lower Tully, Herbert and Burdekin catchments relative to the overall capital and annual operation and maintenance costs of the backbone infrastructure. If implemented it was calculated that the Bradfield Scheme could result in an overall reduction in anthropogenic load of total suspended solids and particulate nitrogen delivered to the Great Barrier Reef of 10 and 8%, respectively. Although there may be some potential for possible ecological benefits in diverting water from north Queensland and strategically releasing it to try and achieve better environmental outcomes for the northern MDB, it would need to be considered against the range of possible ecological impacts both within the northern MDB and within the Bradfield Scheme source catchments. These impacts are likely to be large. The upper limit of the regional benefit from expanded farming would be less than $6.1 billion per year and create up to 11,000 jobs, but this would only occur if: • Horticulture in the vicinity of St George were to expand by an amount equivalent to 30% of the gross value of all fruit, vegetable and nuts currently produced in all of Australia (which could take more than 100 years if this were even possible) • Farmers did not have to pay for water. Most of the new jobs would be for seasonal fruit picking and packing work, which historically has largely been undertaken by foreign and domestic migratory workers from outside the region (so much of the benefit would accrue outside the northern MDB, and partially outside Australia). The fundamental weakness of Bradfield-style schemes are the high costs of diversion infrastructure (without an offsetting locational benefit) and large volumes of expensive water (in excess of what high-value industries could expand to consume and pay for in the medium term). While it may be difficult to find new water development options that are financially viable in their own right, it would at least be possible to find more efficient alternatives that avoid such needlessly high financial losses. Such alternatives could involve strategically planned regionally-distributed additional water storages that could be progressively staged to scale with the demand of high-value water users. This could be done in two ways, depending on whether the priority was to cost-effectively expand local bulk water supply volumes or increase regional water security (or some combination of the two). If the priority were on increasing water volumes to allow local high-value industries to expand, then augmenting regional water storage capacity to capture additional water supplies and meet local demands would likely achieve the highest benefit per million dollars of infrastructure spending. If the priority were on improving local water security and water supply reliability, then drought-related water shortfalls could be mitigated by gradually building additional water supply capacity and by linking relatively smaller pieces of new and existing infrastructure into a number of ‘interconnected regional grids’. However, this may come at the expense of inefficiencies from having infrastructure on standby in non-drought periods during which it would not be used to its full capacity. These alternative options for configuring water infrastructure would be able to meet the water supply and security objectives of Bradfield-style schemes with less risk, lower cost and better matching of water infrastructure development to where demands and opportunities (in terms of where natural resources and other existing infrastructure are located) are already greatest.

Item ID: 75929
Item Type: Report (Report)
Keywords: Northern Australia, water security, Murray-Darling Basin, irrigation, water resource development, inter-basin transfer, dams, diversion infrastructure, horticulture
Related URLs:
Copyright Information: © Commonwealth Scientific and Industrial Research Organisation 2021. To the extent permitted by law, all rights are reserved and no part of this publication covered by copyright may be reproduced or copied in any form or by any means except with the written permission of CSIRO.
Additional Information:

Peer reviewed report

Funders: National Water Grid Authority
Date Deposited: 19 Oct 2022 02:58
FoR Codes: 38 ECONOMICS > 3801 Applied economics > 380105 Environment and resource economics @ 100%
SEO Codes: 18 ENVIRONMENTAL MANAGEMENT > 1803 Fresh, ground and surface water systems and management > 180301 Assessment and management of freshwater ecosystems @ 50%
15 ECONOMIC FRAMEWORK > 1505 Microeconomics > 150510 Production @ 50%
Downloads: Total: 2
More Statistics

Actions (Repository Staff Only)

Item Control Page Item Control Page