Recent results from a randomized wintertime cloud seeding experiment in the Snowy Mountains of Australia

The Snowy Mountains are one of the few regions of Australia that have seasonal snowfall accumulations and weather situations that are suitable for winter cloud seeding. Although Australia has a long history in cloud seeding research, the last major experiment in the Snowy Mountains occurred in the 1950s. In fact, throughout the world randomized experiments to evaluate the effectiveness of wintertime glaciogenic cloud seeding have been quite rare over the past 30 years. The interest in cloud seeding over the Snowy Mountains as a means of increasing snowfall was renewed in the 1980s, prompted by many decades of declining water supplies from snow melt. After several feasibility studies and an Environmental Impact Statement, the Snowy Precipitation Enhancement Research Project (SPERP) was launched in 2004. To date the project, which targets a 1000 km2 area in the Snowy Mountains, has completed four of the six projected winter field campaigns.

The principal evaluation of SPERP will be based on a statistical evaluation of precipitation over the target during randomly selected seeded and non-seeded experimental units. The project is also being continuously assessed through a variety of physical and chemical measurements to validate steps in the conceptual model used by SPERP. One particular aspect, which is unique to a wintertime randomized seeding experiment, is the use of a dual trace chemistry methodology (simultaneous release of ice nucleating silver iodide and non-ice nucleating indium oxide aerosols from all seeding sites). This is employed to verify targeting of the seeding aerosol, to differentiate between ice nucleation and scavenging processes, and to develop a new technique for quantifying the effects of cloud seeding.

This paper describes the design and operational aspects of SPERP and presents results from physical and snow chemistry studies conducted during a 2-day storm period in the 2006 winter field campaign. A unique time series of trace chemical results from the centre of the target area is presented to show the evolution of concentration ratios of seeding and tracer chemical elements corresponding to five separate experimental unit periods. The time series appears to differentiate between seeded and unseeded events. The spatial pattern of the trace chemical concentrations in the snowpack for this storm period is also presented and shows a high percentage of sites being targeted, as well as a lack of targeting at the intended control site.