THE INCORPORATION OF SULFUR-DIOXIDE INTO SNOW AND DEPOSITING ICE.

Abstract

Depth profiles of S(IV) and S(VI) in snow exposed to 20-140 ppbv SO₂ for 6 to 12 hours have been determined in 48 laboratory experiments. Surface deposition velocity (v(d)) averaged 0.06 cm s⁻¹. Well-metamorphosed snow, longer run times, higher SO₂ concentrations and colder snow were associated with lower values of v(d), and vice versa. Melting followed by draining increased v(d) greatly (0.14 cm s⁻¹. Any effect of ozone on SO₂ v(d) was undetectable. Most sulfur in the snow was as S(VI), even without added ozone, indicating the presence of other oxidants, especially in new snow. Four NO₂ deposition experiments (average v(d) = 0.007 cm s⁻¹), and one combined SO₂-NO₂ deposition experiment were conducted. Ozone, sunlight and SO₂ did not enhance NO₂ deposition; NO₂ and sunlight did not enhance SO₂ deposition. The deposition of SO₂ into a snowpack is modelled as an aqueous system, where the liquid water is considered to be present on snow grain surfaces. Gas transport into the snow, air-water partitioning, and aqueous-phase reactions are explicitly considered. Three oxidants (Fe- or Mn-catalyzed O₂, O₂, and H₂O₂) act to convert S(IV) to S(VI), acidify the film, and inhibit further S(IV) uptake. Model calculations illustrate the primary importance of liquid-water mass fraction (X(m)) and the secondary importance of oxidative reactions on SO₂ v(d) to snow. Model and experimental results are similar for assumed X(m) on the order of one percent. Experiments were also conducted on the incorporation of SO₂ into ice depositing from the vapor at -7 and -15°C. Remarkably, SO₂ is captured in deposited ice at concentrations comparable to Henry's Law equilibrium with water at 0°C. Ozone and HCHO appear to inhibit, not enhance, SO₂ capture. An aqueous-film model accounting for the capture of SO₂ by depositing ice was developed. S(IV) concentrations may be enhanced in the liquid-like layer on growing ice surfaces due to solute exclusion from the bulk ice and greatly-retarded diffusional transport from the ice/film interface, leading to significant incorporation into the ice despite low distribution coefficients. SO₂ snow scavenging ratios may be comparable to sulfate scavenging ratios in the remote troposphere.