Recent experiments have sought to determine if hygroscopic seeding of individual convective clouds can improve the clouds precipitation efficiency by enhancing the coalescence process within them (Mather et al., 1997; Silverman and Sukarnjanaset, 2000; Bruintjes et al., 2001).

Seeding at cloud base with hygroscopic material to produce precipitation increases is predicated on the assumption that the rain-producing process evolves in the following manner: (1) the introduction at cloud base of large and giant cloud condensation nuclei (CCN) produced by burning hygroscopic flares in racks mounted to the wings of the seeder aircraft; (2) preferential activation of the larger CCN from the flares, leading to a broadening of the cloud droplet distribution; (3) growth of the large cloud droplets into raindrops via natural coalescence processes, in clouds which could not otherwise have grown raindrops through warm-rain processes; (4) the transport of the raindrops into the supercooled portion of the cloud where the raindrops freeze due to their larger size; (5) invigoration of the cloud due to released latent heat and growth of the frozen drops to large graupel by accretion of the cloud water; and (6) increased radar-estimated rainfall at cloud base and presumably more rainfall at the ground, when the enhanced water mass moves downward through the cloud (Mather et al., 1997).

Planned research with hygroscopic seeding in Texas was conducted in 2005 utilizing the SOAR research aircraft and its crew. Prof. Daniel Rosenfeld and Dr. William Woodley identified a patented means of processing common salt (NaCl) to virtually any desired size as verified after production by analysis with an electron microscope (Figure 1). A desiccant was added to the salt powder during its production to prevent its clumping. The model simulations of Segal et al. (2004) indicated that 3 to 5 microns diameter would be the optimal size range and particles of this size were produced for initial experiments to be conducted in Texas (SPECTRA II).
 

During SPECTRA II Drs. Woodley and Rosenfeld together with the SOAR crew conducted several hygroscopic seeding experiments using milled salt released from an agricultural aircraft. SF6 gas was released in updraft from the cloud-base seeder simultaneous with the release of the hygroscopic salt powder. The gas was detected on subsequent passes, indicating that the aircraft had penetrated the seeded plume. The detection of the gas was successful in part due to a software program that ingests the seeder coordinates and the research aircraft coordinates and directing the research aircraft towards the location of the seeder aircraft (figure 3). Although the data is still being analyzed, it is evident that an alteration of the cloud droplet spectra near and above cloud base, as required by the conceptual model, was found within the seeded plume. During the Texas experiments, such a seeding signature was observed as a broadening of the precipitation particle spectrum as measured in the Cloud Imaging Probe (CIP). This is shown in figure 4. More cases are needed in order to draw any conclusions.
 

References :
Bruintjes, R. T., D. W. Breed, V. Salazar, M. Dixon, T. Kane, G. B. Foote and B. Brown, 2001: Overview and results from the Mexican hygroscopic seeding experiment. Preprints,AMS Symposium on Planned and Inadvertent Weather Modification, Albuquerque NM.

Mather, G. K., D. E. Terblanche, F. E. Steffens, and L. Fletcher, 1997: Results of the South African cloud-seeding experiments using hygroscopic flares. J. Appl. Meteor., 36, 1433-1447.

Silverman, B. A. and W. Sukarnjanaset, 2000: Results of the Thailand warm-cloud hygroscopic seeding experiment. J. Appl. Meteor., 39, 1160-1175.