Report – Water Demand Forecast and Infrastructure Planning Report

Executive Summary

This report assesses the long-term water security of a hypothetical regional water corporation, the Authority, which supplies an inland regional city of about 85,000 residents in south-eastern Australia. It quantifies current demand, projects it to 2044 using Australian Bureau of Statistics population growth assumptions, tests the supply and demand balance against the system secure yield, and appraises a portfolio of demand management, water recycling and new supply options. Modelled average-day demand in the 2024 base year is 21.25 ML per day, or 7,756 ML per year, against a secure yield of 9,000 ML per year, giving comfortable headroom of 1,244 ML per year, or 16.0 per cent of demand. Demand grows with population to 10,238 ML per year by 2044, and the base-case supply and demand balance moves into deficit in 2035. Peak-day demand exceeds the existing treatment plant capacity of 45 ML per day earlier, at about 2032, and the 10 per cent operating headroom target is breached by about 2028, which is the point at which augmentation should be committed given typical infrastructure lead times of five to seven years. A least-cost portfolio that combines a permanent demand management program, leakage reduction and staged surface water augmentation is recommended, supported by a climate-adjusted reassessment of secure yield and a five-yearly adaptive planning cycle consistent with Victorian urban water strategy requirements.

Introduction

Urban water corporations in Australia are required to plan for security decades ahead because the assets that deliver water, storages, treatment plants and trunk mains, take many years to approve, fund and build. Under the National Water Initiative and the Victorian framework, water businesses prepare an Urban Water Strategy every five years that forecasts demand, assesses supply, and identifies the actions needed to keep supply and demand in balance across a 50-year outlook (Department of Energy, Environment and Climate Action 2022). Sound planning is also an economic obligation: Infrastructure Australia (2021) frames water security as a least-cost problem in which demand management and non-build options must be weighed against new infrastructure before capital is committed, and the Water Services Association of Australia benchmarks the sector against this expectation (Water Services Association of Australia 2022).

This report applies that discipline to the Authority. It has three aims: to quantify current demand and project it to 2044; to test the supply and demand balance and identify the deficit year; and to appraise the options for closing the gap. The scope covers the potable urban supply system serving the regional city and its townships; bulk irrigation and rural stock and domestic supply are excluded. All figures are modelled estimates, internally consistent and calibrated against published Australian sector benchmarks rather than any single real utility.

Demand Analysis

Current demand baseline

Average-day demand (ADD) is the product of the connected population and the system-wide per-capita consumption rate, which for the base year is taken as 250 litres per person per day. This figure spans residential use, non-residential and institutional use, and non-revenue water, and sits within the range that the Water Services Association of Australia (2022) reports for regional systems. The base-year calculation is:

• ADD = P × q / 10⁶ = 85,000 × 250 / 1,000,000 = 21.25 ML per day
• Annual average demand = 21.25 × 365 = 7,756 ML per year

Non-revenue water, principally distribution leakage and metering under-registration, is estimated at about 12 per cent of system input. Because it represents demand that can be reduced without affecting customers, it is treated later as a supply option in its own right.

Population and growth assumptions

Population is the dominant demand driver. Growth is projected using a compound annual rate of 1.4 per cent, consistent with Australian Bureau of Statistics projections for growing inland regional centres (Australian Bureau of Statistics 2023). The compound form is used because a regional city accumulates dwellings year on year rather than growing in a straight line. For the end of the horizon:

• P(2044) = 85,000 × (1.014)²⁰ = 85,000 × 1.321 = 112,200 residents

Per-capita consumption trend

Per-capita consumption is held constant at 250 litres per person per day across the horizon. This is a deliberately conservative base case: efficiency gains from appliances and gardens are expected to be offset by warmer conditions and larger dwellings, so a flat rate isolates population as the driver. The saving from an active demand management program is credited separately in the later option appraisal, which is the correct place for a discretionary intervention.

Demand Forecast, 2024 to 2044

Applying the population projection and the constant per-capita rate produces the demand forecast in Table 1. Peak-day demand is estimated by applying a peak-day factor of 1.9 to average-day demand, a value typical of regional Victorian systems where hot, dry summers concentrate garden and evaporative-cooling use into a small number of days. Peak day governs the sizing of treatment and pumping assets, whereas annual demand governs the adequacy of the source. For 2044, peak-day demand is 28.05 × 1.9 = 53.3 ML per day.

Table 1: Modelled water demand forecast for the Authority, 2024 to 2044

Note. Figures are modelled estimates for a hypothetical utility.

Two constraints emerge. The source constraint is annual demand approaching the secure yield, examined next. The treatment constraint arrives sooner: the plant is rated at 45 ML per day, and peak-day demand reaches that ceiling at 2029 + (45 – 43.3) / (46.4 – 43.3) × 5 = 2032. Peak-day capacity is therefore the first physical limit the Authority will encounter, and it must be considered alongside the source balance.

Supply and Demand Balance

The relevant measure of supply is secure yield, the volume the system can reliably deliver each year through a repeat of the worst drought on record without breaching restriction rules (Loucks and van Beek 2017). The Authority’s storages, groundwater bores and bulk entitlement provide a current secure yield of 9,000 ML per year. Headroom is the surplus of secure yield over demand. Table 2 sets the demand forecast against this yield and reports headroom in volume and percentage terms.

Table 2: Supply and demand balance against a secure yield of 9,000 ML per year

Note. Figures are modelled estimates for a hypothetical utility. Values in parentheses denote a deficit.

The balance is worked as headroom = secure yield less demand. In 2039 this is 9,000 – 9,554 = (554) ML per year, a shortfall equal to 5.8 per cent of demand. The base-case deficit year, at which demand first equals secure yield, is found by interpolating between the 2034 and 2039 forecasts: 2034 + (9,000 – 8,915) / (9,554 – 8,915) × 5 = 2034 + 0.7 = 2035. Prudent practice, however, is not to wait for zero headroom. Applying a target operating headroom of 10 per cent, augmentation must be in service before demand reaches 9,000 / 1.10 = 8,182 ML per year, which occurs at 2024 + (8,182 – 7,756) / (8,313 – 7,756) × 5 = 2028. Because major supply schemes take five to seven years to plan, approve and build, the commitment decision must be taken close to the start of the horizon, not near the deficit itself. Figure 1 illustrates the widening gap and the crossing point.

Climate and Drought Risk

The forecast above treats secure yield as fixed, but yield is itself climate-dependent, and this is the largest source of uncertainty in the plan. The Bureau of Meteorology and CSIRO report that cool-season rainfall across southern Australia has declined and that this drying trend is projected to continue, reducing inflows to storages even where annual rainfall changes little (Bureau of Meteorology and CSIRO 2022). Regional projections for south-eastern Australia point to more frequent, more intense droughts and higher evaporative loss from open storages (CSIRO and Bureau of Meteorology 2015). The Millennium Drought of about 2001 to 2009 is the reference event for this system and showed that streamflow can fall far more sharply than rainfall, because dry catchments absorb a greater share of what falls.

The practical implication is that the 9,000 ML per year secure yield may be optimistic under a drying climate. A stress test that reduces yield by 12 per cent to a climate-adjusted 7,920 ML per year brings the balance into deficit almost immediately, since 2029 demand of 8,313 ML per year already exceeds it. Victorian planning guidance requires exactly this kind of climate-adjusted assessment within the Urban Water Strategy, testing supply against a range of future streamflow scenarios rather than the historical record alone (Department of Energy, Environment and Climate Action 2022). Drought risk therefore compresses the timeline: the deficit year is 2035 on historical yield but could arrive within the current decade if a Millennium-scale drought recurs, which strengthens the case for early, staged action.

Supply Augmentation and Demand Management Options

Six options span the three standard levers of demand management, water recycling and new supply. Consistent with least-cost planning, each is appraised on a levelised cost of water (LCOW) basis so that a demand-side saving and a new source are compared on the same measure of dollars per megalitre (Turner et al. 2016; Infrastructure Australia 2021). Table 3 reports indicative yield, capital cost, levelised cost and lead time.

Table 3: Appraisal of supply and demand options (real discount rate 6 per cent)

Note. Figures are modelled estimates for a hypothetical utility.

The levelised cost converts a lump-sum capital cost into an equivalent annual charge and adds operating cost, then divides by annual yield. The annualisation uses a capital recovery factor (CRF): CRF = r(1 + r)ⁿ / ((1 + r)ⁿ – 1). Taking the recycled water scheme as a worked example, with a real discount rate r of 6 per cent and an asset life n of 30 years, (1.06)³⁰ = 5.744, so CRF = (0.06 × 5.744) / (5.744 – 1) = 0.3446 / 4.744 = 0.0727. The annualised capital cost is A$45 million × 0.0727 = A$3.27 million per year. Adding annual operating and maintenance costs of A$0.90 million gives A$4.17 million per year, and dividing by the scheme yield of 800 ML per year gives a levelised cost of A$4.17 million / 800 = A$5,210 per ML, as shown in Table 3.

The appraisal reveals a clear cost order. Demand management and leakage reduction are the cheapest sources of water at A$1,410 and A$1,820 per ML, and they are also the fastest to deliver, which is why least-cost planning gives them first priority (Marsden Jacob Associates 2019). Surface water augmentation offers the largest single block of new yield at a moderate A$3,070 per ML but carries the longest lead time. Recycling and the regional pipeline are the most expensive per megalitre, but their value lies elsewhere: recycled and stormwater sources are largely rainfall-independent and therefore hedge the drought risk identified above, a benefit a simple cost ranking understates. Efficient water pricing that signals scarcity to customers reinforces the demand-side options and improves the return on every supply investment (Grafton and Wheeler 2018; Wheeler 2022). No single option should be selected in isolation; a portfolio that sequences cheap demand-side action first and commits to bulk supply only as far ahead as lead times require is both lower cost and more resilient than a single large scheme (Productivity Commission 2021).

Recommendations

1. Adopt a portfolio rather than a single scheme. Fund the water efficiency and leakage reduction programs now, since together they yield about 1,150 ML per year at the lowest cost and within two to three years, deferring the deficit and reducing the size of any later build.
2. Commit to surface water storage augmentation by about 2028 so that its 1,500 ML per year is in service before the 2035 deficit, allowing for the five to seven year lead time. The combined demand-side and surface water yield of about 2,650 ML per year restores headroom well beyond the 2044 shortfall of 1,238 ML per year.
3. Progress the recycled water and stormwater schemes as the drought hedge, targeting non-potable and industrial customers first, because their rainfall independence protects supply when storages are low even though their unit cost is higher.
4. Reassess secure yield on a climate-adjusted basis using the drying scenarios required under Victorian urban water strategy guidance, and re-run the balance against the lower yield before finalising the capital program.
5. Address the treatment constraint in parallel by staging an uplift of plant capacity above 45 ML per day ahead of the 2032 peak-day limit, so that the source and treatment programs are coordinated.
6. Govern the plan as a five-yearly adaptive cycle, monitoring per-capita consumption, population and streamflow annually and re-forecasting the balance, so that investment is triggered by evidence rather than by a fixed date.

Conclusion

The Authority currently enjoys comfortable water security, with 16.0 per cent headroom over a secure yield of 9,000 ML per year, but that position is temporary. Population growth of 1.4 per cent per year lifts annual demand from 7,756 to 10,238 ML per year across the horizon to 2044, and on historical yield the system enters deficit in 2035. The binding constraints arrive sooner in practice: peak-day demand reaches the treatment plant ceiling at about 2032, the 10 per cent operating headroom target is breached by about 2028, and a recurrence of a Millennium-scale drought could bring the deficit forward into the current decade. The appraisal shows that demand management and leakage reduction are the cheapest and fastest responses and should be funded first, that surface water augmentation must be committed by about 2028, and that recycling has value as a drought hedge beyond its headline cost. Managed as an adaptive, climate-adjusted planning cycle rather than a one-off build, the Authority can maintain water security across the 20-year horizon at the lowest whole-of-system cost.

References

Australian Bureau of Statistics (ABS) 2023, Population Projections, Australia, 2022 (base) to 2071, cat. no. 3222.0, ABS, Canberra.

Bureau of Meteorology (BoM) & CSIRO 2022, State of the Climate 2022, Bureau of Meteorology, Melbourne.

CSIRO & Bureau of Meteorology 2015, Climate Change in Australia: Projections for Australia’s Natural Resource Management Regions, CSIRO, Canberra.

Department of Energy, Environment and Climate Action (DEECA) 2022, Central and Gippsland Region Sustainable Water Strategy, Victorian Government, Melbourne.

Grafton, RQ & Wheeler, SA 2018, ‘Economics of water recovery in the Murray-Darling Basin, Australia’, Annual Review of Resource Economics, vol. 10, pp. 487-510.

Infrastructure Australia 2021, 2021 Australian Infrastructure Plan, Infrastructure Australia, Sydney.

Loucks, DP & van Beek, E 2017, Water Resource Systems Planning and Management: An Introduction to Methods, Models, and Applications, 2nd edn, Springer, Cham.

Marsden Jacob Associates 2019, The economic value of improved water security, report prepared for the Australian Water Association, Marsden Jacob Associates, Melbourne.

Productivity Commission 2021, National Water Reform 2020, Inquiry Report no. 96, Productivity Commission, Canberra.

Turner, A, Fane, S, Kazaglis, A & White, S 2016, Least-cost planning and demand management for the urban water sector, Institute for Sustainable Futures, University of Technology Sydney, Sydney.

Water Services Association of Australia (WSAA) 2022, Urban Water Industry Performance Report 2021-22, WSAA, Melbourne.

Wheeler, SA 2022, ‘Debunking Murray-Darling Basin water trade myths’, Australian Journal of Agricultural and Resource Economics, vol. 66, no. 4, pp. 797-821.