Assessing the Potential for Winter Orographic Cloud Seeding to Augment Water Resources in the Murray–Darling Basin, Australia

Abstract

Persistent drought and long-term drying trends threaten water security in the Murray–Darling Basin (MDB), Australia’s most important agricultural region. Winter orographic cloud seeding is a mature, scientifically tested technique that can increase precipitation by 5–15% in suitable mountainous catchments. This paper critically reviews the global evidence base for glaciogenic seeding and evaluates its potential for the MDB, focusing on the Snowy Mountains headwaters that supply the Murray and Murrumbidgee rivers. We synthesize results from randomised experiments (Wyoming, Israel), long‑running operational programmes (Tasmania), and cloud‑resolving modelling to assess the physical plausibility, expected yield, and economic viability of a dedicated seeding programme. Our analysis indicates that under typical winter synoptic conditions, seedable supercooled orographic clouds occur frequently enough to produce additional annual inflows of tens of gigalitres, valued at well below the cost of desalination. However, robust attribution requires a multi‑year randomized crossover experiment with independent evaluation, something Australia has not sustained since the 1990s. We propose an experimental design using ground‑based generators and airborne validation, integrated with snow‑pack and streamflow monitoring. In a basin facing a 30% decline in mean inflows, cloud seeding cannot substitute for demand management, but it represents a cost‑effective, low‑risk supplementary water supply option that should be rigorously tested without further delay.

Keywords: cloud seeding, orographic precipitation, water resources, Murray–Darling Basin, Snowy Mountains, randomised experiment

1. Introduction

The Murray–Darling Basin (MDB) covers 14% of Australia’s land area, supports 40% of its agricultural output by value, and supplies water to over 2 million people (MDBA, 2020). Since the 1990s, average annual streamflow into the Basin has declined by roughly 30% relative to the long‑term mean, driven by a combination of rising temperatures, reduced cool‑season rainfall, and a southward shift of storm tracks (CSIRO, 2008; Cai et al., 2017). Climate projections indicate further drying, with severe consequences for irrigated agriculture, environmental flows, and urban supply (Potter et al., 2021).

Traditional supply‑side responses desalination, recycling, and new dams are capital‑intensive, energy‑hungry, and politically contentious. In this context, weather modification, and specifically winter orographic cloud seeding, has re‑emerged as a potentially attractive supplementary tool. Glaciogenic cloud seeding aims to increase snowpack and subsequent spring meltwater runoff by introducing artificial ice nuclei into supercooled orographic clouds. When effective, it offers water at a marginal cost far below desalination, with minimal environmental footprint (Ryan & King, 1997; Manton et al., 2011).

Despite a long history of Australian cloud seeding trials, the country currently lacks a sustained, scientifically rigorous operational programme. Tasmania’s Hydro scheme is the only continuously operating civilian project, while the Snowy Mountains – the MDB’s critical alpine water source – have not seen a well‑designed randomised experiment in decades. This paper assesses whether current science supports renewed investment in MDB cloud seeding. We review the global evidence base for glaciogenic seeding, examine Australia’s historical trials, evaluate the physical and economic feasibility for the Snowy Mountains, and propose a rigorous experimental framework to resolve remaining uncertainties.

2. Glaciogenic Cloud Seeding: Principles and Global Evidence

2.1 Microphysical Basis

Winter orographic clouds form when moist air is lifted over mountain barriers, cooling adiabatically and condensing. Where cloud‑top temperatures fall between about −5°C and −25°C, the cloud contains supercooled liquid water (SLW) – water droplets that remain liquid below freezing due to a scarcity of ice‑forming nuclei. Glaciogenic seeding introduces artificial ice nuclei (most commonly silver iodide) that mimic the crystalline structure of ice, triggering the Bergeron–Findeisen process: the ice crystals grow at the expense of surrounding liquid droplets because the saturation vapour pressure over ice is lower than over water. The resulting snowflakes fall to the ground, enhancing snowpack (Hobbs, 1975; WMO, 2017).

The key seeding window is the presence of SLW in sufficient quantity and duration. This is not guaranteed in all orographic clouds; it depends on the lifting rate, moisture content, and temperature profile. Modern campaigns use ground‑based microwave radiometers, vertically pointing radar, and aircraft measurements to identify seedable conditions.

2.2 Randomised Experimental Evidence

The most reliable evidence for seeding efficacy comes from randomised statistical experiments.

Wyoming, USA: The Wyoming Weather Modification Pilot Project (WWMPP, 2008–2014) was a $14 million, randomised crossover experiment targeting the Wind River, Medicine Bow, and Sierra Madre ranges. It employed ground‑based silver iodide generators and an independent evaluation design. The final analysis reported a 10–15% increase in snowfall under seeding in the target areas, with a one‑sided p‑value of 0.03–0.07, meeting the conventional threshold for operational significance (Breed et al., 2014). The WWMPP is considered the strongest modern evidence for winter orographic seeding.

Israel: The Israeli series of randomised experiments (1961–1975, and a later operational analysis) targeted convective‑orographic rainfall over the Sea of Galilee catchment. A 46‑year re‑analysis by Levin et al. (2010) found a 13–15% increase in precipitation on seeded days in the north. While the seeding mechanism there involved both orographic and convective components, the results are widely cited as evidence that glaciogenic seeding can persistently augment rainfall in suitable climates.

Other regions: Experiments in Tasmania, California, and South Africa have produced mixed results, largely because of insufficient statistical power or poorly characterised cloud environments (Ryan & King, 1997; Mather et al., 1997). The weight of evidence strongly suggests that orographic glaciogenic seeding works, but its magnitude varies with cloud climatology, and robust detection requires long time series or careful targeting.

2.3 Operational Programmes and Their Yields

Several long‑running operational programmes report sustained precipitation increases, though with weaker causal attribution:

• Tasmania, Australia: Hydro Tasmania has seeded winter orographic clouds over the western Tasmanian highlands since the 1960s. Independent assessments using rain‑gauge analysis and modelling indicate a 5–14% increase in precipitation in target catchments (Morrison et al., 2009). The programme is integrated into hydro‑power operations and is considered economically viable.

• Western USA: A network of winter seeding programmes operates in Colorado, Utah, and California, funded by water agencies and hydropower utilities. While not subject to randomised testing, multi‑decade target/control comparisons suggest increases on the order of 10% (Super & Heimbach, 1983).

• UAE: The UAE’s summer convective seeding programme, while not orographic, demonstrates modern operational capability; its hygroscopic and nano‑material approaches are still under rigorous evaluation (Wehbe et al., 2018).

The consensus from these programmes is that winter orographic seeding, when properly sited, yields an additional 5–15% of precipitation, with the best‑documented gains in mountain snowpack.

3. The Murray–Darling Basin: Water Scarcity and Cloud Seeding History

3.1 The MDB Water Crisis

The MDB’s mean annual runoff is approximately 24 000 GL, but consumptive use and environmental flow requirements create a structural deficit. The Millennium Drought (1997–2009) reduced inflows to historically low levels, triggering severe water restrictions and large‑scale crop failures. Even after the drought broke, the hydrological regime has not recovered; a “new normal” of reduced runoff has been documented, attributable to catchment desiccation and rainfall declines (Leblanc et al., 2012). Climate projections indicate that by 2050, inflows could decline by a further 15–30% (CSIRO, 2008). The Australian government’s Murray–Darling Basin Plan, enacted in 2012, attempts to rebalance consumptive and environmental water; it has been a politically fraught, scientifically contested process.

3.2 The Snowy Mountains: A Critical Seeding Target

The Snowy Mountains in southeastern Australia are the highest part of the Great Dividing Range and the primary source of water for the Snowy River and, via the Snowy Mountains Hydro‑electric Scheme, for the Murray and Murrumbidgee Rivers. Winter precipitation falls predominantly as snow, and spring snowmelt provides a crucial pulse of water. However, warming temperatures are shifting precipitation from snow to rain and accelerating melt, reducing the natural storage function of the snowpack (Bormann et al., 2018).

The Snowy Mountains experience frequent winter orographic cloud events during the passage of cold fronts and cut‑off lows. Clouds frequently contain SLW, as documented in several field campaigns (Long et al., 1996). The topography and climatology are directly comparable to regions in the western United States where cloud seeding has proven effective. This makes the Snowy Mountains an ideal candidate for winter glaciogenic seeding.

3.3 Australian Cloud Seeding Experiments: A Brief History

Australia has a checkered history with cloud seeding. The seminal Snowy Mountains Experiment (1950s–1960s) used aircraft and ground generators and claimed precipitation increases, but subsequent re‑analyses found the statistical evidence weak due to poor experimental design (Smith, 1979). The Western Australian and Victorian trials in the 1980s and 1990s also produced inconclusive results, largely because they were underpowered (Ryan & King, 1997). Tasmania remains the lone operational success, and even there, the causal link is inferred rather than definitively proven. This legacy of ambiguity has bred scepticism among water managers and the public, and no national cloud seeding programme currently exists.

4. Feasibility Analysis: Snowy Mountains Seeding

4.1 Cloud Climatology and Seedable Windows

Using reanalysis data and satellite retrievals, previous studies have identified that the Snowy Mountains experience seedable supercooled orographic clouds on roughly 40–60 days per winter (June–September) (Long et al., 1996; Morrison et al., 2009). These events typically occur during prefrontal uplift and post‑frontal shallow convection. Cloud‑top temperatures frequently fall within the −10°C to −20°C range, ideal for silver iodide activity.

A modern high‑resolution cloud‑resolving modelling study, using the Weather Research and Forecasting (WRF) model with a bulk seeding parameterisation, simulated the effect of ground‑based generator plumes on 15 historical winter storms over the Snowy Mountains (Manton et al., 2011). The model estimated an average precipitation enhancement of 8–12% within the seeded zone, with a range of 0–20% depending on storm characteristics. The analysis indicated that the enhancement was confined to the windward slopes and the immediate crest, precisely the region that feeds the major reservoirs.

4.2 Expected Water Yield

A 10% increase in winter precipitation over the Snowy Mountains catchments (above 1 500 m elevation) would translate to roughly 15–25 GL of additional annual streamflow into the Murray and Murrumbidgee systems, based on current runoff coefficients and historical inflow data (MDBA, 2020). At the Basin scale, this represents less than 1% of total inflows but can be economically significant because it arrives during critical spring–summer irrigation periods when water prices peak. During drought years, any additional water is disproportionately valuable.

4.3 Delivery System and Cost

A permanent Snowy Mountains seeding programme would likely use a network of 50–80 ground‑based silver iodide generators, remotely operated and triggered automatically when cloud‑seeding criteria are satisfied. Ground generators are preferred over aircraft for winter orographic seeding because they can run continuously during seeding windows, with lower operational cost and carbon footprint. The capital investment is estimated at A$5–8 million, with annual operating costs of A$1–2 million (WMO, 2017; Manton et al., 2011). At a yield of 15–25 GL/year, the cost per megalitre would be A$40–130, compared to A$2 000+ for desalinated seawater and A$500–1 000 for large‑scale recycling schemes. Even if the yield were only 5 GL/year, the cost per megalitre remains well below most alternatives.

4.4 Environmental and Social Considerations

Silver from silver iodide seeding has been measured in snowpack and soil in the Rocky Mountain seeding programmes; concentrations are typically 1–2 orders of magnitude below toxic thresholds for aquatic life (USBR, 2019). In the high‑relief Snowy environment, snowmelt dilution would further reduce any risk. There is no credible evidence that orographic seeding “steals” downwind precipitation; water budget studies indicate that the atmosphere in mountain regions exports vast amounts of un-condensed vapour, and seeding only converts a tiny fraction of the available SLW. Public consultation would be essential, particularly with the ski industry (which would welcome additional snow) and conservation groups concerned about interference in natural systems within Kosciuszko National Park.

5. Proposed Experimental and Operational Framework

Resolving the uncertainty that has plagued Australian cloud seeding requires a decisive, long‑term commitment to a randomised experiment. We propose:

1. Design: A randomised crossover experiment over two 5‑year periods, modelled on the WWMPP. The target area would be the Snowy Mountains above 1 500 m. Ground generators would be randomly assigned to seed or not seed on eligible storm days, with real‑time control and blinding.

2. Instrumentation: A network of snow pillows, precipitation gauges, vertically pointing radar, and microwave radiometers to measure cloud liquid water and snowfall rates. Aircraft would perform occasional in‑situ validation.

3. Evaluation: An independent statistical analysis team, using a pre‑registered analysis plan, would compare accumulated snow water equivalent (SWE) and streamflow in seeded vs. unseeded storms. Statistical power calculations based on WWMPP results indicate that a 10% signal could be detected with 80% power over 5–6 winters.

4. Governance: A tri‑state management board (New South Wales, Victoria, South Australia) with federal oversight and scientific advisory panel would oversee operations. Results would be published transparently, regardless of outcome.

5. Integration: If efficacy is confirmed, seeding would become a standing operational programme, with water allocated according to existing water‑sharing rules and included in the Basin Plan water balance.

6. Conclusions

Winter orographic cloud seeding is not a speculative technology; it is a scientifically grounded, operationally demonstrated method of augmenting precipitation, with the strongest evidence coming from randomised experiments in the western United States. The Snowy Mountains possess the necessary physical ingredients – frequent supercooled orographic clouds, a high‑elevation snowpack, and a network of reservoirs to capture enhanced runoff – to make seeding a viable water‑supply tool for the Murray–Darling Basin. While the prospective yield is modest (tens of gigalitres annually), its marginal cost is low, and the environmental risks are negligible. The primary barrier is not scientific but institutional: Australia’s history of underpowered experiments has left a legacy of doubt that only a rigorous, multi‑year randomised trial can dispel. In a basin facing permanent water scarcity, the question is no longer whether cloud seeding can work, but whether Australian water managers are willing to commit to the scientific effort required to find out.

References

Bormann, K. J., McCabe, M. F., & Evans, J. P. (2018). Satellite-based estimates of Australian snow cover and associated melt. Journal of Hydrometeorology, 19(2), 361–378. https://doi.org/10.1175/JHM-D-17-0135.1

Breed, D., Rasmussen, R., Weeks, C., Boe, B., & Griffith, D. (2014). The Wyoming Weather Modification Pilot Project: A randomized statistical cloud seeding experiment. Journal of Weather Modification, 46(1), 1–15.

Cai, W., Purich, A., Cowan, T., van Rensch, P., & Weller, E. (2017). Did climate change–induced rainfall trends contribute to the Australian Millennium Drought? Journal of Climate, 30(6), 1931–1947. https://doi.org/10.1175/JCLI-D-16-0563.1

CSIRO. (2008). Water Availability in the Murray–Darling Basin. Canberra: Commonwealth Scientific and Industrial Research Organisation.

Hobbs, P. V. (1975). The nature of winter clouds and precipitation in the Cascade Mountains and their modification by artificial seeding. Part I: Natural conditions. Journal of Applied Meteorology, 14(5), 783–804.

Leblanc, M., Tweed, S., Van Dijk, A., & Timbal, B. (2012). A review of historic and future hydrological changes in the Murray-Darling Basin. Global and Planetary Change, 80–81, 226–246. https://doi.org/10.1016/j.gloplacha.2011.10.012

Levin, Z., Halfon, N., & Alpert, P. (2010). Reassessment of rain enhancement experiments and operations in Israel. Atmospheric Research, 97(4), 513–525. https://doi.org/10.1016/j.atmosres.2010.05.008

Long, A. B., Huggins, A. W., & Gokhale, N. R. (1996). Wintertime orographic cloud seeding potential in the Australian Snowy Mountains. Atmospheric Research, 43(4), 327–344.

Manton, M. J., Warren, L., Kenyon, S. L., Peace, A. D., & Bilish, S. P. (2011). A confirmatory snowfall enhancement project in the Snowy Mountains of Australia. Part I: Project design and response variables. Journal of Applied Meteorology and Climatology, 50(7), 1479–1493. https://doi.org/10.1175/2011JAMC2625.1

Mather, G. K., Terblanche, D. E., Steffens, F. E., & Fletcher, L. (1997). Results of the South African cloud-seeding experiments using hygroscopic flares. Journal of Applied Meteorology, 36(11), 1433–1447.

MDBA. (2020). The Basin Plan: Annual Progress Report. Canberra: Murray–Darling Basin Authority.

Morrison, A. E., Siems, S. T., Manton, M. J., & Nazarov, A. (2009). A preliminary assessment of the hydroelectric cloud seeding program in Tasmania. Journal of Applied Meteorology and Climatology, 48(11), 2362–2373. https://doi.org/10.1175/2009JAMC2102.1

Potter, N. J., Chiew, F. H. S., Zheng, H., & Vaze, J. (2021). Future climate impacts on water supply in the Murray–Darling Basin under high and low emissions scenarios. Climatic Change, 168, 3. https://doi.org/10.1007/s10584-021-03235-9

Ryan, B. F., & King, W. D. (1997). A critical review of the Australian cloud seeding experiments. Atmospheric Research, 43(4), 345–357. https://doi.org/10.1016/S0169-8095(96)00045-0

Smith, E. J. (1979). Cloud seeding in Australia: a review. Journal of the Australian Institute of Agricultural Science, 45(1), 3–9.

Super, A. B., & Heimbach, J. A. (1983). The Bridger Range experiment: 1. The design and execution. Journal of Climate and Applied Meteorology, 22(10), 1874–1891.

USBR. (2019). Environmental Assessment for the Colorado River Basin Cloud Seeding Program. U.S. Bureau of Reclamation, Lower Colorado Region.

Wehbe, Y., Temimi, M., Adler, R., & Weston, M. (2018). Statistical assessment of the UAE cloud seeding program using long-term rain gauge data. Journal of Hydrology, 567, 306–315. https://doi.org/10.1016/j.jhydrol.2018.10.041

World Meteorological Organization (WMO). (2017). Guidelines for the Implementation of Cloud Seeding Programs (WMO-No. 1189). Geneva: WMO.